Int Arch Occup Environ Health (2004) 77: 39–46
DOI 10.1007/s00420-003-0476-z
O R I GI N A L A R T IC L E
Ari Kaukiainen Æ Tapio Vehmas Æ Kaarina Rantala
Markku Nurminen Æ Rami Martikainen
Helena Taskinen
Results of common laboratory tests in solvent-exposed workers
Received: 22 December 2002 / Accepted: 20 August 2003 / Published online: 5 November 2003
Ó Springer-Verlag 2003
Abstract Objectives: The screening and identification of
occupational liver or other organ-system injury related
to long-term, low-level solvent exposure are difficult in
clinical practice. We studied the feasibility of the use
of common laboratory tests combined with a detailed
exposure history. Methods: The relationships between
laboratory tests and exposure to organic solvents were
studied in regression modelling adjusted to age, alco-
hol consumption, gender and body mass index (BMI).
The subjects were 29 solvent-exposed workers and 19
referents. Laboratory tests included serum alanine
aminotransferase (ALT), aspartate aminotransferase
(AST), gamma-glutamyl transferase (GGT), carbohy-
drate-deficient transferrin (CDT), alkaline phosphatase
(ALP), creatinine, cholesterol, HDL-cholesterol, tri-
glycerides, blood glucose and total and conjugated
bilirubin. Positive hepatitis serology, systemic diseases
or medications with known hepatic effects and current
pregnancy were exclusion criteria. The main exposures
of each subject were identified. Current solvent expo-
sure status, exposure during the past 3 months, expo-
sure during the past 5 and 10 years, and total life-time
exposure were recorded. Results: AST (P=0.0031),
ALT (P=0.0015) and cholesterol (P=0.0110) corre-
lated positively with cumulative solvent exposure in the
past 5 years, total bilirubin with current exposure
(P=0.0380), and glucose with exposure in the past 5
(P<0.0001) and 10 (P=0.0003) years. Triglycerides
correlated positively with exposure in the past 5
(P=0.0025) and 10 (P=0.0059) years and with life-
time exposure (P=0.0005). Creatinine correlated neg-
A. Kaukiainen (&) Æ T. Vehmas Æ K. Rantala Æ M. Nurminen
R. Martikainen Æ H. Taskinen
The Finnish Institute of Occupational Health (FIOH),
Topeliuksenkatu 41 a A,
00250 Helsinki, Finland
E-mail: Ari.Kaukiainen@ttl.fi
Tel.: +358-9-47472284
Fax: +358-9-2412414
atively with exposure in the past 10 years (P=0.0300)
and life-time exposure (P=0.0005). Most laboratory
values were within the normal range. Conclu-
sions: These results suggest a multi-system health effect
of solvents. The laboratory data had some similarities
with those in the metabolic syndrome. The screening
and diagnostics of solvent-related conditions should be
based on a thorough work history and a set of care-
fully selected laboratory tests. No single test seems
sufficient for this purpose.
Keywords Glucose Æ Creatinine Æ Triglycerides Æ
Liver Æ Solvents
Introduction
Organic solvents are widely used in industry, and despite
their declining exposure levels, health hazards still exist.
Various examples of potentially dangerous tasks include
the degreasing of metal, paint manufacture, painting and
lacquering, and dry-cleaning. The route of exposure of
volatile petrochemical substances is usually through the
airways or skin; thus, many organ systems may be af-
fected simultaneously. Long-term, low-level or interact-
ing occupational exposures may cause liver injury that is
difficult to identify (Chen et al. 1997). In addition to the
liver (Brautbar and Williams 2002; Warnes et al. 2000),
deleterious effects have been found in the nervous system
(White and Proctor 1997), renal system, cardiovascular
system, skin, mucous membranes, and on reproductive
and developmental health; even carcinogenicity has been
described or suspected (Baker 1994; Xiao and Levin
2000).
After the occupational exposure history has been
registered, common laboratory tests are usually the first-
line diagnostic tool when deleterious solvent effects are
suspected. Liver enzymes are often determined in rou-
tine occupational health examinations, for the screening
of latent liver disease, alcohol abuse or adverse liver
effects caused by solvent exposure.
40

The sensitivity and specificity of liver-function tests
are considered to be low in chronic low-grade or latent
liver dysfunction caused by chemicals (Chen et al. 1997).
The most common cause of abnormal liver-function
test results in the general population in western countries
is hepatic steatosis associated with obesity. If metabolic,
autoimmune and viral disorders have been excluded,
occupational exposure to chemicals needs to be carefully
considered (Warnes et al. 2000).
We performed laboratory tests on workers with a
detailed history of solvent exposure and on a reference
group in order to identify some common laboratory tests
that might reveal the possible effects of current or past
occupational exposure to solvents.
Subjects and methods
The exposed group consisted of 29 workers who had been con-
secutively referred to the Department of Occupational Medicine
at the Finnish Institute of Occupational Health (FIOH) by
occupational health physicians between January 2000 and Janu-
ary 2001 because of suspected solvent-related chronic toxic
encephalopathy or polyneuropathy. Exclusion criteria included
systemic diseases or medication with known hepatic effects, po-
sitive hepatitis serology and current pregnancy. One subject was
excluded because of both type 1 diabetes mellitus and positive
hepatitis-B serology; another because of an existing diagnosis of
alcoholism; and a third due to an exposure history consisting
only of a past 1-day poisoning from an unknown mixture of
chemicals. All the invited subjects participated. The study group
consisted of three women and 23 men (mean age 51, range 29–66
years). The reference group consisted of 19 volunteers, who were
employees of the FIOH and had had no or minimal solvent
exposure (four women and 15 men, mean age 45, range 30–57
years). One referent was excluded from the regression model
studying the relationship between solvent exposure and serum
triglycerides, due to a distinctly increased value of serum trigly-
cerides (7.0 mmol/l). All the subjects were of the same ethnic
origin.
The workers
occupations included spray painter, printer, con-
struction painter, factory worker, laboratory worker, glass-fibre
laminator, carpet layer, carpenter, surface finisher and maintenance
worker (Table 1). The reference group comprised volunteers who
were employees of the Institute, with white-collar workers pre-
dominating.
The study design was approved by the local ethics committee.
All subjects participated voluntarily and gave their written, in-
formed consent.
The study protocol included a questionnaire and laboratory
tests. Health history, including medication, work and exposure,
was enquired about, as was alcohol consumption. Body weight and
height were asked for, so that the body mass index (BMI) could be
calculated. The workers were interviewed and examined by an
occupational health physician as part of the examination for the
suspected occupational disease.
Alcohol consumption was determined with three core-AUDIT
(alcohol use disorders identification test) questions that included
the types of beverages as well as quantities and frequencies of
consumption (Bush et al. 1998; Gordon et al. 2001). The alcohol
consumption index was calculated as the average number of drinks
per month (one drink equals 12.5 g absolute alcohol).

Table 1 Workers by occupation and solvent exposure (OELY total
life-time cumulative solvent exposure in OEL years, Past 10y
exposure during the past 10 years in OELYs,Past 5yexposure
during the past 5 years in OELYs, Past 3m exposure during the
past 3 months: 0 = none, 1 = slight, 2 = moderate, 3 = high;
main solvents:T toluene,X xylenes,S styrene,WS ‘‘white spirit’’;
aliphatic mixtures: Ac acetates, Al alcohols, HH halogenated
hydrocarbons,n-H n-hexane,G glycol ethers, K ketones, TF tetra-
hydrofuran)

Worker Exposure

Occupation Age OELY Past 10y Past 5y Past 3m Current Main solvents

Laminator (reinforced plastics industry) 28 5.0 4.6 3.2 0 no S

Spray painter (cars) 51 8.0 1.5 0.4 0 no T,X,WS,Ac,Al,G,S
Laboratory worker (metallurgy) 66 6.4 0.2 0 0 no HH,Al,K
Spray painter (railway carriages) 53 8.3 1.0 0.3 0 no T,X,WS,Ac,G,K
Spray painter (car/dockyard) 53 9.9 1.6 0.3 0 no T,X,WS
Printing-trade worker (offset) 48 6.6 2.1 0.9 0 no WS,Al
Car painter/repair-shop worker 60 10.7 2.8 1.3 2 yes T,X,Ac, S
Painter (metal constructions/shipyard) 44 13.7 1.6 0.5 0 no T,X,WS,Ac
Factory worker (industrial coatings) 52 6.9 1.6 0.2 1 yes T,X,Ac,Al,HH,K
Spray painter (wood) 53 9.6 4.0 2.0 1 no T,X,Ac,Al,G,K
Spray painter (wood/metal) 50 11.9 2.7 1.0 0 no T,X,WS,Ac,K
Spray painter (cars/metal) 54 15.4 4.2 0.4 1 yes T,X,WS,Ac,Al,G
Spray painter (metal) 57 16.9 2.5 1.1 1 yes T,X,Al,G
Spray painter/shoemaker 54 7.0 0.6 0.6 0 no T,X,n-H,K
Construction painter 53 4.5 0.2 0.2 0 no WS
Factory worker (plastics industry) 51 9.9 0 0 0 no HH
Laboratory worker (chemistry) 47 1.2 1.2 0.7 1 yes T,HH,K
Printer (flexography) 45 13.3 3.7 1.9 2 yes Ac,Al,K
Printer (screen printing) 55 10.4 3.1 1.0 0 no T,X,WS,Ac,Al,K
Carpenter/surface finisher 48 3.5 3.1 1.2 0 no WS
Carpet layer 51 7.2 2.1 0.9 0 no T,n-H,HH,K
Construction painter 64 8.9 1.0 0.4 0 no T,X,WS,Ac,Al,HH
Surface finisher (wood) 36 4.0 2.3 0.8 1 yes T,WS,Ac,Al,G
Laminator/spray painter (metal) 43 9.1 2.2 1.1 2 yes S,X
Carpet layer 51 7.2 2.1 0.9 1 yes T,WS,Ac,n-H,TF,K
Maintenance worker (gravure printing) 59 6.7 3.0 1.5 2 yes T
 41

Laboratory tests were contacted for further information on solvent exposure. In

Blood samples from both the workers and reference group were
obtained from the cubital vein, in the morning after they had fasted
for 12 h.
Standard laboratory methods were used. Serum levels of ala-
nine aminotransferase (ALT), aspartate aminotransferase (AST),
gamma-glutamyl transferase (GGT), alkaline phosphatase (ALP),
creatinine, cholesterol, HDL-cholesterol, triglycerides and glucose
were determined at the Institute
s laboratory on the day of blood
sampling by use of a selective chemistry analyser, KonePro
(Konelab, Thermo Clinical Labsystems Oy, Finland). ALT, AST
and GGT were determined according to the International Feder-
ation of Clinical Chemistry and Laboratory Medicine (IFCC)
method. The test for creatinine was based on the method of the
Jaffe reaction, and glucose was determined by an enzymatic
hexokinase method.
The serum sample was sent on the same day to an accredited
commercial laboratory (United Laboratories, Helsinki, Finland)
for determination of total and conjugated bilirubin by the serum
diazo-reaction method, carbohydrate-deficient transferrin (CDT)
by the transferrin isoform-separation radioimmunoassay method,
and for serological tests by the enzyme immunoassay method. The
serological tests included: hepatitis-B virus, core antibodies; hepa-
titis-B virus, surface antigen; hepatitis-C virus, antibodies. For the
reference range of tests, see Table 2. The normal value for each
serological test was negative.
Exposure assessment
Exposure was assessed by an industrial hygienist (K.R.) together
with a specialist in occupational medicine (A.K.), according to the
clinical practice at the Institute. The estimation was based on the
work description (e.g. total duration in occupation, description of
work tasks, including solvent exposure, ventilation, and use of
personal protective devices). The worker was asked to report the
solvents or the trade names of solvent-containing products and to
bring the material data safety sheets. If possible, the workplaces
some cases, the results of hygiene measurements were available on
the workplaces. If no measurements were available, previous ones
from similar workplaces, gathered over the years in the Register of
Industrial Hygiene Measurements of the FIOH were used (Tossa-
vainen and Jaakkola 1994).
Exposure assessment of the reference group was based on the
questionnaire on current and previous occupations, as well as on
duration of employment, with special emphasis on work tasks that
possibly involved exposure to solvents.
Each work task was assessed separately. Each subject
s main
exposures were identified, and current exposure status was classi-
fied as ‘‘exposed’’ or ‘‘not exposed’’. The total lifetime cumulative
solvent exposure and the exposure during the past 5 and 10 years
were expressed as occupational exposure limit (OEL) years.
Occupational exposure limit years (OELYs) is the hypothetical
number of years during which the solvent exposure is at the level of
the Finnish Occupational Exposure Limit. One OELY is equivalent
to working 8 h per day for 1 year with the solvent(s) at the Finnish
OEL. Finnish OELs are prepared in accordance with the limits and
recommendations set by Nordic, European and international ex-
pert groups, and the European Commission (Ministry of Social
Affairs and Health 2001).
The general form of cumulative solvent exposure of each sub-
ject is given by:
OELY ¼ Ri Ci ti
where Ci is the average solvent level in the ith work task expressed
as a percentage of the Finnish OEL, and ti is total duration in years
of the ith work task.
Exposure during the past 3 months was classified as 0 = none,
1 = slight, 2 = moderate, 3 = high, where high exposure was
equivalent to cumulative exposure above 3/12 OELYs.
The lifetime cumulative exposure in OELYs ranged from 1.2–
16.9 for the workers (Table 1) and from 0–0.5 for the reference
group. Most workers had been exposed to a mixture of solvents.
The most common solvents were toluene and xylene. Also, ali-
phatic mixtures (‘‘white spirit’’), acetates, alcohols, and ketones
were common, but only four workers were exposed to halogenated

Table 2 Comparison of exposed workers and reference group. Means of background data and the laboratory tests. Significances between
the differences and coefficients of determination (R2) of the adjusted model.F female, M male

Factor Workers Reference group Reference range P R2

Mean (range) Mean (range) Unadjusted Adjustedb

Covariates
Age 51.4 (29–66) years 44.9 (30–57) years
Alcohol consumption indexa 22.8 (0–100) 20.1 (0–56)
Body mass index 26.6 (18.5–33.7) 25.5 (18.5–35.9)
Gender (F/M) 3/ 23 4/ 15
Blood tests
AST 31.4 (8–55) 28.3 (18–47) F: 10–35 U/l, M: 10–50 U/l 0.2863 0.4676 0.20
ALT 40.1 (6–118) 31.7 (14–69) F: 10–35 U/l, M: 10–50 U/l 0.2203 0.1136 0.26
GGT 44.2 (13–144) 37.0 (14–158) F: 5–50 U/l, M: 5–80 U/l 0.4487 0.4941 0.13
CDT 14.4 (8.0–33.0) 16.7 (8.0–34.0) F: <28 U/l, M: <20 U/l 0.2461 0.1335 0.50
AST/ALT 1.09 (0.47–2.50) 0.97 (0.64–1.35) 0.3876 0.9815 0.21
Glucose 5.23 (4.40–6.20) 4.82 (4.0–5.60) 3.5–5.6 mmol/l 0.0039 0.0183 0.19
Creatinine 84.4 (58–102) 91.7 (67–117) <115 U/l 0.0311 0.0050 0.40
ALP 159.8 (84–249) 152.2 (95–234) 60–275 U/l 0.5082 0.8837 0.08
Cholesterol 5.78 (4.1–8.8) 5.69 (4.0–7.8) <5.0 mmol/l 0.7766 0.6220 0.10
HDL-cholesterol 1.47 (0.81–2.62) 1.46 (0.86–2.77) >0.9 mmol/l 0.9550 0.4363 0.52
Triglycerides 1.47 (0.67–3.45) 0.94 (0.46–1.40) 0.50–1.70 mmol/l 0.0018 0.0018 0.30
Bilirubin, total 11.5 (4–33.0) 13.4 (5–26) 2–20 lmol/l 0.3455 0.8454 0.24
Bilirubin, conjugated 2.30 (1–8) 3.08 (1–6) <6 lmol/l 0.2662 0.7671 0.16
Serological tests Negative Negative Negative
a
Calculated from the three core-AUDIT questions as average number of portions per month (one portion equals 12.5 g absolute alcohol)
b
Adjusted for age, gender, BMI and use of alcohol
42

hydrocarbons. Many workers had already been transferred from
solvent-exposed tasks when the examinations started at the FIOH.
Only ten workers were currently exposed to solvents, whereas none
in the reference group had been exposed.
Statistical methods
To study the relationship between solvent exposures and the results
of selected laboratory tests, we fitted a generalised linear regression
model to describe the outcome data that were assumed to follow a
gaussian (normal) distribution. The assumption of normality was
checked by graphical inspection. We used a scaled deviance statistic
(chi-squared distributed) to test the significance of the model
parameters. In all models, we adjusted for the potential con-
founding effect of the covariates (age, gender, BMI, use of alcohol).
We also examined the interaction of solvent exposure at work with
the person
s alcohol consumption by entering their product term
into the model. For the computation we used the general linear
model (glm) function implemented in the S-PLUS system (Venables
and Ripley 2001, chap. 7). We used a smooth nonparametric
function to depict graphically the relation between exposure and
the laboratory values (Venables and Ripley 2001, chap. 9).
Results
Table 2 shows the background data and laboratory test
results of the exposed workers and the reference group.
The groups differed significantly in the mean levels of
blood glucose, serum creatinine and triglycerides. Ta-
ble 3 shows significant relationships between solvent
exposure and laboratory results, both in the whole study
group and in the sole group of the solvent-exposed
workers. Liver enzyme activity (AST, ALT) and serum
cholesterol were related to exposure during the past 5
years, triglycerides to both the past 5 and 10 years.
Triglycerides also correlated with the total lifetime
cumulative solvent exposure. The only laboratory
parameter that was related to current exposure was total
bilirubin. The AST/ALT ratio was not elevated. Blood
glucose level correlated with exposure during the past 5
years (Fig 1) and the past 10 years. An inverted U-
shaped relationship between blood glucose level and the
total lifetime cumulative solvent exposure given in OE-
LYs was noted (Fig 2). The serum creatinine level was
negatively associated with long-term exposure (Fig 3).
We also studied the interaction, i.e. the interdepen-
dence of co-action and the modification of relationship,
between personal alcohol consumption and occupa-
tional solvent exposure in terms of OELYs. In the case
of triglycerides and glucose, the co-action was antago-
nistic; in other words, the influence of the factors on
each other was preventive, so that one blocked the effect
of the other. In contrast, alcohol and solvent exposure
had a synergistic effect on CDT, whereas their solo ac-
tions did not.
Discussion
We used common laboratory tests together with detailed
exposure history in order to study the feasibility of using
these tests with long-term, low-level exposure to organic
solvents. Several positive relationships between the lab-
oratory tests studied and solvent exposure were found.
The serum creatinine level correlated negatively with
long-term exposure to organic solvents.
Exposure to hydrocarbon solvents reported in the
literature decreased fourfold in 1960–1998 (Caldwell
et al. 2000). In our material, major current solvent
exposure was rare. Also, the declining number of refer-
rals compared with those of the 1980s restricted the
number of workers recruited during the enrolment per-
iod. Thus, it is not surprising that many of our subjects
with suspected solvent-related nervous system disorder
received most of their cumulative solvent exposure be-
fore the 1990s and had a long period of work and
exposure. This is reflected by the relatively high mean
age (51 years) of the workers in our study.
Neurotoxic, hepatoxic and other solvent-related
health effects have been described (Baker 1994; Xiao and
Levin 2000). The final clinical assessment of hepato-
toxicity and solvents should take into account synergism
with medication, drugs of use and abuse, alcohol, age,
and nutrition (Brautbar and Williams 2002). In the
present study these factors were considered, together
with viral disorders and systemic diseases with known
hepatic effects. Possible differences in dietary habits,
socioeconomic status and personal living habits between

Table 3 Significant
relationships between solvent Factor Relationship All subjects R2 Solvent-exposed
exposures and laboratory to exposure workers only
variables and coefficients of
determination (R2). All P P R2
regression models adjusted for
age, gender, BMI and use of AST Past 5y + 0.0031 0.33 0.0046 0.36
alcohol. + Positive ALT Past 5y + 0.0015 0.37 0.0769 0.37
relationship,) negative Glucose Past 5y + <0.0001 0.34 0.0017 0.51
relationship, Past 5y exposure Past 10y + 0.0003 0.29 0.0946 0.39
during the past 5 years in Creatinine Past 10y ) 0.0300 0.36 0.9453 0.29
OELYs, Past 10y exposure OELY ) 0.0005 0.45 0.0441 0.39
during the past 10 years in Cholesterol Past 5y + 0.0110 0.21 0.0053 0.40
OELYs, Current current Bilirubin, total Current + 0.0380 0.31 0.0125 0.41
exposure Triglycerides Past 5y + 0.0025 0.29 0.5367 0.20
Past 10y + 0.0059 0.27 0.7236 0.19
OELY + 0.0005 0.33 0.1331 0.25

Fig. 1 Relationship between blood glucose level and past 5-year
cumulative solvent exposure
the index and reference groups could have caused some
residual confounding. BMI, almost identical between the
exposed and control groups, was an indicator of the
nutritional status. This suggests that the magnitude of
the possible dietary difference between lipid or carbo-
hydrate intake was not major. Also, the main results
remained when the sole workers group was studied,
which further suggests the minimal confounding role of
dietary aspects in this study.
We chose workers of the Institute as controls in order
to have a detailed exposure history, to exclude any
hazardous exposure. Besides direct comparison, we used
general linear models adjusted for covariates to avoid
the possible bias caused by differences between the
groups.
Alcohol consumption and viral hepatitis were con-
sidered carefully, because they have a synergistic action
in promoting the development and progression of liver
disease (Gao 2002). Serology other than that of hepatitis
Fig. 2 Relationship between blood glucose level and total lifetime
cumulative solvent exposure
43
Fig. 3 Relationship between serum creatinine level and total
lifetime cumulative solvent exposure
B and C was not tested, because in Finland, as in many
industrialised countries, the clinical significance of other
hepatotoxic viruses is minor in working population.
Hepatitis A was not tested for, because the incidence in
Finland is low: only 51 reported cases during 2000
(National Public Health Institute 2001). Also, hepatitis
A normally resolves without any consequences after the
acute infection (Goeser 2002).
The findings in the current study are not likely to be
alcohol-related, because all regression models were ad-
justed for alcohol consumption. Mean self-reported
alcohol consumption was low, and the mean AST/ALT
ratio was near 1 (an elevated ratio above 1.5 is a well-
documented indicator of alcohol-related liver disease).
Conversely, an elevated ALT/AST ratio has been re-
ported in association with toluene exposure (Guzelian
et al. 1988). Neither was the mean CDT elevated in our
study. CDT is considered to be an alcohol-specific liver-
function test (Behrens et al. 1988; Bell et al. 1993). Body
weight rather than alcohol consumption may be the
major factor in determining the serum level of liver en-
zymes (Lee et al. 2001).
In our material, exposure to aliphatic halogenated
hydrocarbons, generally considered to be the most
hepatotoxic solvents (Warnes et al. 2000), was rare.
However, the usual workplace solvent exposure is to
mixtures, and even mixtures of solvents such as toluene
and xylene have been associated with liver perturbation
(Brautbar and Williams 2002). Toluene and xylene were
also the most common solvents in the present study.
Despite a number of studies on occupational expo-
sure that affects the liver, the feasibility of the use of
liver-function tests remains unclear, especially with
chronic, low-level mixed solvent exposure. Lundqvist
et al. (1999) reviewed approximately 20 such articles that
were published in the 1980s and 1990s and noted, in five
articles, raised ALT activities and, in two articles, ele-
vated AST levels. Increased serum bile acid concentra-
tions were also observed occasionally. Many of the tests
44
used in our study, such as those for AST, ALT or GGT,
reflect current hepatocellular injury, and would be ex-
pected to be sensitive to the effects of ongoing exposure.
The results of these routine liver-function tests usually
show, also, an increase in cirrhosis and other chronic
liver diseases. A combination of liver-function tests is
the approach most likely to reveal even chronic effects
on the liver (Lundberg et al. 1994). Serum hepatic
transaminases ALT and AST reflect liver injury associ-
ated with necrosis, whereas ALP and GGT point to
hepatic cholestasis (Brodkin et al. 2001). Even solvent
exposure, dating back several years, could have effects
on the liver (Lundqvist et al. 1999). As heavy, previous
exposure to solvents, with no or minimal current expo-
sure, may elicit changes in routine liver-function tests
indicating mild chronic liver injury (Lundberg et al.
1994), we scrutinised the past exposure history as thor-
oughly as possible. Our results reflect, especially, the
chronic effects of solvent exposure.
Hepatic transaminases alone may underestimate
hepatotoxic effects in populations exposed to solvents.
Hepatic imaging with ultrasound has been proposed as a
possible sensitive marker for pre-clinical effects of
chronic solvent exposure (Brodkin et al. 1995). A study
with type 2 diabetic patients suggested that non-invasive
measurement of liver fat content (proton spectroscopy)
is a more sensitive index of steatosis than the elevation
of ALT (Ryysy et al. 2000).
Blood glucose levels were within the normal range in
the study group, but they correlated positively with
solvent exposure. Only one exposed subject had im-
paired fasting glucose (fasting plasma glucose concen-
tration 6.1–6.9 mmol/l). Recent studies have associated
non-alcoholic fatty liver (NAFL) with insulin resistance
(Marchesini et al. 1999) and with the metabolic syn-
drome (Lonardo et al. 1995). The mechanisms linking
insulin resistance and fatty liver are unclear (Ryysy et al.
2000). NAFL, in the presence of normoglycaemia and
normal or moderately increased body weight, is char-
acterised by clinical and laboratory data similar to those
found in diabetes and obesity, and may be considered an
additional feature of the metabolic syndrome with spe-
cific hepatic insulin resistance (Marchesini et al. 2001).
The metabolic syndrome probably has a multi-factorial
aetiology, including genetic and environmental factors
(Poulsen et al. 2001). Liver steatosis has been associated
with occupational exposure to volatile petrochemical
substances (Cotrim et al. 1999), and organic solvents
may play a role in the development of fatty-liver disease
(Lundqvist et al. 1999). Trichloroethylene exposure has
been associated with a triphasic response in insulin lev-
els, which depended on the duration of exposure (Goh
et al. 1998). The inverted U-shaped relationship found in
our study, between blood glucose level and total lifetime
cumulative solvent exposure, is in accordance with this
previous finding.
Serum cholesterol and serum triglycerides correlated
positively with solvent exposure in the current study.
Metabolic abnormalities characteristic of the metabolic
syndrome are: elevated levels of plasma glucose, serum
insulin and triglycerides and lower levels of HDL-cho-
lesterol (Ribeiro-Filho et al. 2001). The prevalence of
NAFL with hyperlipidaemic disorders is highly variable,
ranging from 20–92% (Falck-Ytter et al. 2001). NAFL
also occurs in normal-weight men without diabetes and
hyperlipidaemia (Bacon et al. 1994). The mean BMI of
our workers (26.6 kg/m2) was beyond the normal limits
but was only slightly higher than that of the referents.
The characteristics of the metabolic syndrome, in addi-
tion to hyperlipidaemia, are obesity, defined as a high
BMI and/or high waist–hip ratio and increased blood
pressure (Groop and Orho-Melander 2001). Workers
occupationally exposed to high concentrations of ben-
zene and xylene had an increased prevalence of arterial
hypertension (Kotseva and Popov 1998). Both pre-
eclampsia and hypertension alone were associated with
solvent exposure in pregnant women (Eskenazi et al.
1988).
The only laboratory parameter that correlated posi-
tively with current exposure was total bilirubin, whereas
no significant association was seen between conjugated
bilirubin and exposure. Other studies have reported a
higher prevalence of conjugated bilirubin in workers
exposed to solvents than in controls (Tomei et al. 1999;
Brodkin et al. 2001).
We found a negative relationship between serum
creatinine level and solvent exposure. Creatinine for-
mation has a direct relationship with muscle mass; it
increases with skeletal-muscle necrosis and atrophy,
such as trauma or rapidly progressing diseases (Woo
and Henry 1996). A slow toxic process that gradually
diminishes the muscle mass might lower the level of
serum creatinine. Both alcohol (Lang et al. 2001; Preedy
et al. 2001) and occupational solvents (Pedersen et al.
1980; Kossler 1991) may cause myopathy, but a con-
founding factor may also be present. Workers with a
long history of solvent exposure may feel sick and reduce
their normal activity, especially during their leisure time.
This could cause secondary muscle atrophy.
Significant associations between solvent exposure and
the levels of liver enzymes, serum glucose, serum creat-
inine, bilirubin, cholesterol and triglycerides suggest that
solvents have multi-system effects (Fig 4). Liver toxicity
caused by solvent exposure could lead to liver fatty
degeneration and, consequently, to a multi-system met-
abolic disturbance, resembling the metabolic syndrome.
Concluding remarks: Solvent hepatotoxicity, in terms
of liver fatty degeneration, might be the primary phe-
nomenon, resulting in laboratory data that resemble, in
some respects, those found in the metabolic syndrome.
In this syndrome, liver fatty degeneration has repeatedly
been found, although its role as a causative factor is not
clear. Most laboratory values were within the normal
range, and the associations found are, thus, mainly of
epidemiological interest. The diagnostics of individuals
with solvent exposure should be based on a thorough
work history and a set of carefully selected laboratory
tests. No single test seems sufficient for this purpose.

Fig. 4 Proposed mechanism to
explain the changes in
laboratory parameters
References
Bacon BR, Farahvash MJ, Janney CG, Neuschwander-Tetri BA
(1994) Nonalcoholic steatohepatitis: an expanded clinical en-
tity. Gastroenterology 107:1103–1109
Baker EL (1994) A review of recent research on health effects of
human occupational exposure to organic solvents. J Occup Med
36:1079–1092
Behrens UJ, Worner TM, Lieber CS (1988) Changes in carbohy-
drate-deficient transferrin levels after alcohol withdrawal.
Alcohol Clin Exp Res 12:539–544
Bell H, Tallaksen C, Sjaheim T, Weberg R, Raknerud N, Orjasa-
eter H, Try K, Haug E (1993) Serum carbohydrate-deficient
transferrin as a marker of alcohol consumption in patients with
chronic liver diseases. Alcohol Clin Exp Res 17:246–252
Brautbar N, Williams J 2nd (2002) Industrial solvents and liver
toxicity: risk assessment, risk factors and mechanisms. Int J
Hyg Environ Health 205:479–491
Brodkin CA, Daniell W, Checkoway H, Echeverria D, Johnson J,
Wang K, Sohaey R, Green D, Redlich C, Gretch D (1995)
Hepatic ultrasonic changes in workers exposed to perchloro-
ethylene. Occup Environ Med 52:679–685
Brodkin CA, Moon JD, Camp J, Echeverria D, Redlich CA,
Willson RA, Checkoway H (2001) Serum hepatic biochemical
activity in two populations of workers exposed to styrene. Oc-
cup Environ Med 58:95–102
Bush K, Kivlahan DR, McDonell MB, Fihn SD, Bradley KA
(1998) The AUDIT alcohol consumption questions (AUDIT-
C): an effective brief screening test for problem drinking. Arch
Intern Med 158:1789–1795
Caldwell DJ, Armstrong TW, Barone NJ, Suder JA, Evans MJ
(2000) Hydrocarbon solvent exposure data: compilation and
analysis of the literature. AIHAJ 61:881–894
Chen J-D, Wang J-D, Tsai S-Y, Chao W-I (1997) Effects of
occupational and nonoccupational factors on liver function
tests in workers exposed to solvent mixtures. Arch Environ
Health 52:270–274
Cotrim HP, Andrade ZA, Parana R, Portugal M, Lyra LG, Freitas
LA (1999) Nonalcoholic steatohepatitis: a toxic liver disease in
industrial workers. Liver 19:299–304
Eskenazi B, Bracken MB, Holford TR, Grady J (1988) Exposure to
organic solvents and hypertensive disorders of pregnancy. Am J
Ind Med 14:177–188
Falck-Ytter Y, Younossi ZM, Marchesini G, McCullough AJ
(2001) Clinical features and natural history of nonalcoholic
steatosis syndromes. Semin Liver Dis 21:17–26
Gao B (2002). Interaction of alcohol and hepatitis viral proteins:
implication in synergistic effect of alcohol drinking and viral
hepatitis on liver injury. Alcohol 27:69–72
Goeser T (2002) Clinical aspects of the different forms of viral
hepatitis and their treatment. Semin Thromb Hemost 28 [Suppl
1]:37–46
Goh VH, Chia SE, Ong CN (1998) Effects of chronic exposure to
low doses of trichloroethylene on steroid hormone and insulin
levels in normal men. Environ Health Perspect 106:41–44
45
Gordon AJ, Maisto SA, McNeil M, Kraemer KL, Conigliaro RL,
Kelley ME, Conigliaro J (2001) Three questions can detect
hazardous drinkers. J Fam Pract 50:313–320
Groop L, Orho-Melander M (2001) The dysmetabolic syndrome. J
Intern Med 250:105–120
Guzelian P, Mills S, Fallon HJ (1988) Liver structure and function
in print workers exposed to toluene. J Occup Med 30:791–796
Kossler F (1991) Trichloroethylene interactions with muscle cells. J
Appl Toxicol 11:189–194
Kotseva K, Popov T (1998) Study of cardiovascular effects of
occupational exposure to organic solvents. Int Arch Occup
Environ Health 71 [Suppl]:S87–91
Lang CH, Kimball SR, Frost RA, Vary TC (2001) Alcohol
myopathy: impairment of protein synthesis and translation
initiation. Int J Biochem Cell Biol 33:457–473
Lee DH, Ha MH, Christiani DC (2001) Body weight, alcohol
consumption and liver enzyme activity—a 4-year follow-up
study. Int Journal Epidemiol 30:766–770
Lonardo A, Bellini M, Tondelli E, Frazzoni M, Grisendi A, Pul-
virenti M, Della Casa G (1995) Nonalcoholic steatohepatitis
and the ‘‘bright liver syndrome’’: should a recently expanded
clinical entity be further expanded? (letter) Am J Gastroenterol
90:2072–2073
Lundberg I, Nise G, Hedenborg G, Högberg M, Vesterberg O
(1994) Liver function tests and urinary albumin in house
painters with previous heavy exposure to organic solvents.
Occup Environ Med 51:347–353
Lundqvist G, Flodin U, Axelson O (1999) A case–control study of
fatty liver disease and organic solvent exposure. Am J Ind Med
35:132–136
Marchesini G, Brizi M, Morselli-Labate AM, et al (1999) Associ-
ation of nonalcoholic fatty liver disease with insulin resistance.
Am J Med 107:450–455
Marchesini G, Brizi M, Bianchi G, Tomassetti S, Bugianesi E,
Lenzi M, McCullough AJ, Natale S, Forlani G, Melchionda N
(2001) Nonalcoholic fatty liver disease: a feature of the meta-
bolic syndrome. Diabetes 50:1844–1850
Ministry of Social Affairs and Health (2001) Occupational expo-
sure limits 2000 (in Finnish). HTP-arvot 2000. Sosiaali- ja
Terveysministeriö. Työsuojelusäädöksiä 2000:3. Tampere
National Public Health Institute (2001) KTL B 10/2001. Infectious
diseases in Finland 2000. The National Infectious Diseases
Register, http://www.ktl.fi/ttr/tt_en_2000.pdf
Pedersen LM, Nygaard E, Nielsen OS, Saltin B (1980) Solvent-
induced occupational myopathy. J Occup Med 22:603–606
Poulsen P, Vaag A, Kyvik K, Beck-Nielsen H (2001) Genetic versus
environmental aetiology of the metabolic syndrome among
male and female twins. Diabetologia 44:537–543
Preedy VR, Paice A, Mantle D, Dhillon AS, Palmer TN, Peters TJ
(2001) Alcoholic myopathy: biochemical mechanisms. Drug
Alcohol Depend 63:199–205
Ribeiro-Filho FF, Faria AN, Kohlmann O Jr, Ajzen S, Ribeiro
AB, Zanella MT, Ferreira SR (2001) Ultrasonography for the
evaluation of visceral fat and cardiovascular risk. Hypertension
38:713–717
46
Ryysy L, Häkkinen A-M, Goto T, Vehkavaara S, Westerbacka J,
Halavaara J, Yki-Järvinen H (2000) Hepatic fat content and
insulin action on free fatty acids and glucose metabolism rather
than insulin absorption are associated with insulin requirements
during insulin therapy in type 2 diabetic patients. Diabetes
49:749–758
Tomei F, Giuntoli P, Biagi M, Baccolo TP, Tomao E, Rosati MV
(1999) Liver damage among shoe repairers. Am J Ind Med
36:541–547
Tossavainen A, Jaakkola J (1994) Occupational exposure to
chemical agents in Finland. Appl Occup Environ Hyg 9:28–31
Venables WN, Ripley BD (2001) Modern applied statistics with S-
PLUS, 3rd edn. Springer, New York
Warnes TW, Jain SK, Smith A (2000) Hepatotoxic effects of
workplace exposures. In: Baxter PJ, Adams PH, Tar-Ching A,
Cockcroft A, Harrington JM (eds) Hunter
s diseases of occu-
pations, 9th edn. Arnold, London
White RF, Proctor SP (1997) Solvents and neurotoxicity. Lancet
349:1239–1243
Woo J, Henry JB (1996) Metabolic intermediates and inorganic
ions. In: Henry JB Clinical diagnosis and management by lab-
oratory methods. Saunders, Philadelphia, pp 162–193
Xiao JQ, Levin SM (2000) The diagnosis and management of
solvent-related disorders. Am J Ind Med 37:44–61