Intemational Archives of

Int Arch Occup Environ Health ( 1987) 59:281-294 { 1 { (ia 11 3 I dend

© Springer-Verlag 1987

Toluene metabolism during exposureto varying concentrations combined with exercise

Jesper Baelum l, Martin Dossing 2, Steen Honore Hansen 3,Gunnar R Lundqvistl , and Niels Trolle Andersen 4

Institute of Hygiene, University of Aarhus, DK-8000 Arhus C, Denmark2 Department of Medicine F, Copenhagen County Hospital, Gentofte,DK-2900 Hellerup, Denmark3 Royal Danish School of Pharmacy, Universitetsparken, DK-2100 Copenhagen, Denmark4 Department of Theoretical Statistics, Institute of Mathematics, University of Aarhus,DK-8000 Arhus C, Denmark

Summary The urinary excretion of hippuric acid (HA) and ortho-cresol(O-cr) in man was measured in two studies of 7-h exposure to toluene in aclimate chamber, either constant concentration of 100 ppm or varying con-centrations containing peaks of 300 ppm but with a time-weighted average of100 ppm In Study A, four males were exposed to clean air and to constantand varying concentrations of toluene in combination with rest and with100 W exercise in 140 min Exercise increased end exposure excretion rate ofHA and O-cr by 47 and 114 %, respectively After exposure, all excess HAwas excreted within 4 h, while O-cr was eliminated with a half life of about3 h Alveolar air concentration of toluene varied between 21 and 31 ppmduring constant exposure and between 13 and 57 ppm during varying expo-sure, but no difference in mean alveolar toluene concentration or in metabo-lite excretion was seen between the exposure schedules In Study B, 32males and 39 females aged between 31 and 50 years were exposed once toeither clean air, constant or varying concentrations of toluene Backgroundexcretion rate of HA was 0 97 ± 0 75 mg/min ( 1 25 + 1 05 g/g creatinine) androse to 3 74 + 1 40 mg/min ( 3 90 + 1 85 g/g cr) during the last 3 h of expo-sure to 100 ppm toluene The corresponding figures for O-cr were 0 05 +0.05 gg/min ( 0 08 + O 14 mg/g cr), and 2 04 + 0 84 g/min ( 2 05 + 1 18 mg/gcr) The individual creatinine excretion rate was considerably influenced bysex, body weight and smoking habits, thus influencing the metabolite con-centration standardised in relation to creatinine It is concluded that bothmetabolites are estimates of toluene exposure O-cr is more specific thanHA, but the individual variation in excretion of both metabolites is large,

Offprint requests to: J Belum at the above address

J Baelum et al.

and when implementing either of them as biological exposure indices, the in-fluence of sex, body size, age as well as consumption of tobacco and alcoholhas to be considered.

Key words: Toluene Exercise Varying concentrations Metabolites -Creatinine

Introduction

At work places, the concentrations of organic solvents in the breathing zone areusually quite variable, showing short peaks of high concentrations depending onthe work process (Ovrum et al 1978 ; Kjxergaard et al 1985) However, hygienicstandards are based on time-weighted averages during an 8-h work day, whichis believed to be representative for the body burden of exposure (ACGIH 84-85).

In addition to the inspiratory concentration, the uptake of solvents is deter-mined by the physical activity and other individual factors (Astrand et al 1972 ;Veulemans and Masschelein 1978 ; Carlsson 1982) Biological monitoring hastherefore been suggested as a better estimate of the individual body burdenthan environmental measurements, and for some solvents urinary metaboliteexcretion has been recommended by ACGIH ( 1984) and WHO ( 1981) Alveo-lar air or blood solvent concentrations are mostly used in experimental studies,but hygienic standards have also been proposed (ACGIH 1984).

Two metabolites of toluene have been the target of investigation Hippuricacid (HA) is the main metabolite, and biological monitoring programs usingthis metabolite have been suggested (ACGIH 1984 ; WHO 1981), but the dis-advantage of HA is the lack of specificity (Andersson et al 1985 ; D O ssing et al.1983) Only 0 1 % of toluene is metabolised to ortho-cresol (O-cr), but this per-centage is rather stable, and the background level of O-cr is low (D O ssing et al.1983).

The present studies were set up in order to investigate the influence of peakconcentrations and physical exercise on the body burden of toluene during con-trolled exposure In Study A the time course of alveolar concentrations of to-luene and urinary HA and O-cr excretion during and after exposure was studiedin detail, while in Study B the individual excretion of HA and O-cr was investi-gated in a larger group of subjects.

Material and methods

Subjects

Study A Four healthy males aged 21 to 25 years, body weight 63-78 kg, participated Beforeentering the study, they went through a medical examination and were tested with electrocar-diographic recordings during rest and on an ergometer cycle with loads up to 200 W.

Study B Seventy-one males and females aged 31 to 50 years were included Only personswithout lung, heart or other disabling diseases were allowed to participate Additional criteria

282

Toluene metabolism during exposure to varying concentrations

Table 1 Data of the subjects in Study B The figures are mean values (range in brackets) Onedrink = 12 g pure alcohol

Males Females

No subjects 32 39

Age (years) 38 1 ( 31-50) 39 8 ( 31-50)Height (cm) 176 ( 160-188) 165 ( 153-181)

Weight (kg) 76 ( 59-93) 61 ( 44-80)No smokers 10 18

Daily cigarette consumption of smokers 6 7 ( 2-20) 10 4 ( 1-25)

Daily alcohol consumption (drinks) 1 1 ( 0-3) 0 9 ( 0-3)

Alcohol consumption the day before 0 6 ( 0-4) 0 7 ( 0-3)investigation (drinks)

Work during exposure (no subjects) 50 W 0 275 W 4 23

100 W 28 14

of exclusion were occupational exposure to solvents, consumption of drugs, and a daily intakeof more than three alcoholic drinks ( 36 g pure alcohol).

Before entering the study, all persons went through a medical examination including heartand lung auscultation, measurement of arterial blood pressure, and a lung function test (Vit-alograph, Birmingham, UK) Finally, electrocardiogrammes were recorded at rest and duringexercise on ergometer cycle The work load was increased in steps of 25 W up to 100 W in10 min Only persons with all test results within normal limits were included in the studygroup.

By a randomization stratified according to sex and age the study group was divided intothree groups, which were exposed to constant toluene concentration (Group 1), varying con-centration of toluene (Group 2), and clean air (Group 0), respectively The personal data areshown in Table 1.

Exposure conditions

Both studies took place in the climate chamber at the Institute of Hygiene, University ofAarhus (Andersen et al 1983) The air temperature was 21 0 °C ± O 3 C (mean ± 1 SD), andthe relative humidity was 37 % ± 2 % Fresh air was supplied at a rate of 1245 m 3 ± 23 m 3 h-1

corresponding to an air renewal of 15 ± 0 3 h-1.

Fig 1 Time course of toluene concentration inair during varying exposure The period is re-peated 14 times during exposure The numbersshow mean and standard deviation of the concen-trations during Study B (P = Peak values, Co-15= average concentrations during the first 15 minof the period, and Co-30 = average concentrationduring the whole period)

minutes

283

A

J Baelum et al.

Table 2 Exposure design in Study A W denotes eight periods of 10 min and two 30-minperiods of 100 W exercise on an ergometer cycle 0 denotes rest On Day 1 subjects performedwork either in the morning or in the afternoon

Day Exposure Subject

1 2 3 4

1 Clean air O/W O/W W/O W/O2 Constant W W O O3 Varying W O W O4 Varying O W O W5 Constant O O W W

Two schedules of toluene exposure were set up One was a constant concentration of100 ppm ( 4 1 rmol/l, 375 mg x m 3), and the other was varying concentrations containing apeak every 30 min during the exposure The time course of a 30-min period is shown in Fig 1.The peak concentrations reached 300 ppm, the average concentration during the whole periodwas 100 ppm, and during the 15 min with the highest concentration it was 150 ppm The expo-sure was in accordance with the short term exposure limits recommended by ACGIH ( 1984-85).

To achieve these concentration profiles, an injection system using two different size spraynozzles was constructed Analytical grade toluene preheated to 95 °C was directed via apneumatic valve to one of the nozzles and injected into the recirculation channel of thechamber Toluene concentration in the chamber was continuously measured by a flame-ioni-zation detector (Bendix, model 8401) calibrated against a standard gas ( 102 ppm toluene insynthetic dry air, AGA, Stockholm) A microcomputer sampled the detector signal and oper-ated the valves supplying toluene In the recirculation channel, an explosive vapour detectorwas installed, and the supply of toluene was automatically stopped if explosive vapour concen-tration was detected, if the ventilation was disrupted, or if the toluene concentration in thechamber exceeded 400 ppm These alarms were never activated during the studies.

Experimental design

Study A: The four subjects were exposed for five consecutive days (see Table 2) Each daystarted at 0800 h with a one-hour acclimatization period in clean air followed by 7 h of expo-sure until 1600 h On the first day the exposure was to clean air (control), while toluene wassupplied to the air on the following days On Days 2 and 5 the concentration of toluene was aconstant 98 ± 6 ppm, while on Day 3 and 4 toluene concentrations varyied, with a time-weigh-ted average of 102 and 106 ppm, respectively, and with peaks of 314 ± 41 ppm On two of thefour days with toluene exposure, the subjects performed physical work of 100 W on an er-gometer cycle in eight periods lasting 10 min, and two periods of 30 min On the days withpeak concentrations, the work sessions were performed during the peaks.

Study B: Each person was exposed once to either a constant or a fluctuating concentration oftoluene or to clean air The exposure lasted 7 h preceeded by a one-hour acclimatizationperiod Three or four persons were exposed at a time During exposure the physical work onan ergometer cycle was performed in three 15-min periods, two periods within the first 2 5 hof exposure, and one after 4 h The work sessions were placed during the peaks of the fluctuat-ing concentrations, and the intensity was 50, 75 or 100 W individualized in order to avoid workexceeding 60 % of maximal aerobic capacity computed from the results of the pre-examination(Astrand 1977).

During exposure smoking was prohibited, and no alcohol, tea or coffee was served InStudy A smoking and intake of alcohol were prohibited during the whole period of investiga-tion.

284

Toluene metabolism during exposure to varying concentrations

Measurements

Alveolar toluene concentrations

Alveolar air samples were taken in Study A using a two-bag technique and analysed by aphotoionisation detector (D O ssing et al 1984) Samples were collected before, during, andafter exposure.

Urinary samples

Study A Quantitative sampling of urine was carried out during nine periods: 0900 to 1100,1100 to 1300, 1300 to 1500, 1500 to 1600, 1600 to 1800, 1800 to 2000, 2000 to 2200, 2200 to0700, and 0700 to 0900 on each of the five consecutive days.

Study B The time of the last micturition before entering was registered, and quantitative sam-pling was carried out in the period until 0900, from 0900 to 1300 and from 1300 to 1600 Fur-thermore, two spot samples were taken at 2200 and at 0700 the next morning These two sam-ples were returned by mail within 3 d, and all samples were stored at -200C until analysis.

All samples were analysed for HA and O-cr using a high performance liquid chromatog-raphy technique described earlier (Hansen and D O ssing 1982) Creatinine concentration wasmeasured by the modified Jaffe method (Bonsnes and Tanssky 1945).

The stability of the metabolites, when stored at room temperature, was tested No signifi-cant change in either HA or O-cr concentrations was seen in samples stored in closed contain-ers for up to 14 d.

Statistical procedures

Study A An analysis of variance according to a factorial design was used (Armitage 1971).The factors were person, exposure and work level The F-test was used and probability levelsbelow five percent were regarded as significant.

Study B The distribution of each variable was examined graphically, and the values weretransformed using the natural logarithm, when necessary, to stabilize the variance Thegroups were compared using a standard analysis of variance, and the influence of the averagetoluene concentration, the day of exposure, work level, sex, age, body weight, smoking habitsand alcohol consumption was determined.

Pre-exposure levels for all three groups were compared, while the samples during andafter exposure were analysed separately for Group 0 and for the Groups 1 and 2.

All calculations were carried out using the program pack GENSTAT (Numeric Al-gorithms Group 1980) The level of significance was five percent.

Results

Toluene concentrations in alveolar air

Figure 2 shows the concentrations measured during and after 30 min of work Itcan be seen that a work load of 100 W increases the alveolar concentration by47 % (SD = 5 %) and the maximal alveolar concentration is 28 % (SD = 18 %)higher during work than in the next period at rest During fluctuating concen-trations, the alveolar toluene concentration follows the inspiratory concentra-tion with only a slight dampening, reaching maximal concentrations of57 ± 7 ppm compared with the maximal concentrations of 31 ± 3 ppm duringconstant exposure However, the average alveolar concentrations during these

285

J Belum et al.

minutes

Fig 2 Alveolar toluene concentration(upper figure) and alveolar/ambient airfraction of toluene (lower figure) duringa one-hour period including 30 min ofexercise in Study A (" ) constanttoluene exposure, ( ) varying ex-posure Each curve shows the mean andstandard deviation of four subjects.

0 10 20 30 40 50 60

minutes

one-hour periods were not different ( 26 7 ± 3 4 and 26 6 + 3 4 ppm, respec-tively).

After cessation of exposure, the alveolar concentration declined accordingto a two phase exponential function reaching 1 8 + O 6 ppm after 45 min Thenext morning alveolar toluene concentration in all subjects was below the detec-tion limit ( 0 5 ppm).

Metabolites

Study A The excretion rates during and after exposure are shown in Fig 3.The excretion rate of HA rises gradually during the exposure The increase

is steeper on days with work than on days with rest, with a 47 ± 4 7 % higherrate at the end of exposure, and a 24 + 2 2 % higher average rate during the 7-hexposure ( 11 9 ± 1 9 and 9 5 + 2 4 mmol, respectively) There was no differencein excretion rates between days with constant level and with varying concentra-tions.

After exposure the excretion rate of HA declined sharply, almost reachingcontrol level 4 h after leaving the chamber The 24-h excretion of HA was 21 2 +3.4 mmol and 19 1 + 3 6 mmol at days with and without work (t= 1 73, P> 0 10).In comparison, the 24-h excretion during clean air exposure was 6 7 ± 3 1 mmol.

It can be seen that the O-cr excretion rates increase steadily during exposureand decline exponentially after exposure In all sampling periods, physical ac-

60

286

E

Toluene metabolism during exposure to varying concentrations

50

40

30

C.

20

10

o 2 4 67 9 11 13

a~ O hours

70

60

50

40

E

Fig 3 Urinary excretion rate of hippuric acidand ortho-cresol during and after 7-h exposurein Study A Each curve shows the mean andstandard deviation of four subjects The blackbar below the abscissa denotes the exposureperiod

30

20

10

o 2 4 6 7 9 11 13

a~v hours

tivity gives rise to a higher O-cr rate, and the difference ( 114 %) is largest at theend of exposure (Fig 3) The amount of O-cr excreted during the 7-h exposureon days with exercise was 10 9 + 6 6 gmol and on days with rest 6 3 + 1 2 gmol,while the corresponding 24-h excretions were 24 3 ± 9 6 and 13 6 + 7 2 plmol,respectively No difference between constant and varying concentration was re-corded (P > O 5).

After discontinuation of exposure, the elimination constants of O-cr on daysof work and without work were 0 25 ± 0 04 h 1 and 0 21 + 0 05 h 1, respectively(P < 0 01), corresponding to half lives of 2 7 and 3 3 h.

Study B The main results of HA and O-cr are shown in Fig 4 and in Table 3.

Hippuric acid

Pre-exposure levels of HA did not differ between the groups The excretionrate was 5 4 + 4 2 gmol/min ( 0 97 + 0 75 mg/min) (mean ± SD) and was not in-

D

22 24

O-CRESOLexerciserest clean air

22 24

287

J Balum et al.

0 4 7 13 22

hours

Fig 4 Urinary concentration of hippuric acidand orthocresol standardised in relation tocreatinine before, during, and after 7-h exposureto 100 ppm toluene (n = 47) and clean air (n = 24)in Study B Each curve shows the mean ± onestandard deviation The black bar below theabscissa denotes the exposure period

0 4 7 13 22

hours

fluenced by any of the personal variables During clean air exposure, no changein excretion rate was observed, but at the end of exposure smokers had a higherrate than non-smokers ( 7 4 + 4 0 and 4 2 3 1 gmol/min, respectively, P<0.05), and females tended to have lower HA rates than males This differencewas significant, when smoking habits were included in the analysis.

During toluene exposure, HA excretion rates increased from the first tothe second sampling period The average excretion rate during the first 4 h was12.7 + 4 7 mol/min ( 2 27 O 84 mg/min), and in the last 3 h it was20.9 + 7 9 gmol/min ( 3 74 + 1 40 mg/min) The part of HA arising from tolueneexposure is calculated as (H Atoluene H Aclean air)/H Atoluene (H Atoluene and

H Aclean air are the amounts of HA excreted in the period in the groups exposedto toluene and clean air, respectively) This is 70 % during the 7-h exposureperiod and 75 % during the last 3 h.

3.0

2.0

E

1.0

0.0

3.0

2.0

1.0

0.0

288

^"

Toluene metabolism during exposure to varying concentrations

C t O C ON

C o C6 6 vr u

-C C C 9

00 r t rl r 3

e, 00 ' CC *

o N r C C

C) C C)=-

C C N C

q C C CI CO

C r O O C0 O N C 00 C

l CI Cl C

m¢ x x

o C

I

c N k C

X X

I +

X

t CC Co+ I

X n

Co

CrC 14

x

r-C 4

N \ C 9 C Cq Nl

o

o~ ~~~~~~~~~~~t

ZC)

E > S v Z m~ O

289

c)

E

Q

'-E

0

E

I

_

zs

00

0

-3P0E

a

V)

UC)

-oE

or-2CCC)

r Q

o) n

= c L0

O

C

C Co

~C)

C

E

o 2

.

C)

C

0-e

o 11

50

c

o 3

.b

1

C )

C)

ii

o O

o O

.DOC)

o °

._ O

.

J Baelum et al.

No difference between the groups exposed to constant and to varying con-centration was observed, but males had a 30 % higher average excretion ratethan females The rate was correlated to body weight and to work level, whichexplained the sex difference The creatinine standardised concentration of HAafter exposure showed that almost all excess HA due to toluene inhalation wasexcreted within 6 h of cessation of exposure.

Ortho-cresol

Average O-cr pre-exposure rate was 0 5 ± 0 5 nmol/min ( 0 05 ± 0 05 gtg/min).The rates were much higher in smokers than in non-smokers, as only 29 % ofnon-smokers had values above the detection limit ( 0 02 pmol/l), but the corre-sponding figure among smokers was 88 % The excretion rate increased withdaily tobacco consumption and with age Accordingly, O-cr rates in smokersfell almost to the level of non-smokers at the end of clean air exposure, followedby a rise to pre-exposure levels in the evening, when smoking was allowed.

During the 7-h toluene exposure, the excretion rate rose to 7 9 ± 4 4 nmol/min ( 0 85 + 0 48 ltg/min) in the first 4 h and to 18 9 ± 7 8 nmol/min( 2.04 ± 0 84 pg/min) in the last 3 h The fraction of O-cr arising from toluene inthis period is 97 % if calculated as for HA No difference between persons ex-posed to constant and variable concentrations was seen, but within the groupsexcretion rate increased with age and work load The pre-exposure differencebetween smokers and non-smokers disappeared during exposure.

The time course of the O-cr concentration standardised in relation to urinarycreatinine followed the excretion rate, but the values during toluene exposurewere negatively correlated with body weight, and females had 40 % higher val-ues than males (see Table 3).

After exposure the concentration of O-cr declined to a next morning level of17 % of the end exposure level, and 4 9 times higher than the pre-exposurelevel In this last sample, the difference between smokers and non-smokersagain became apparent The elimination rate was not related to any of the per-sonal variables.

Creatinine excretion

Personal factors influenced differently the excretion rates and the concentra-tions standardised in relation to creatinine of HA and O-cr Therefore, theexcretion rates of creatinine before and during exposure were calculated.

In Study A, a difference in the individual creatinine excretion rate betweenthe four subjects was seen, but there was neither a systematic diurnal variationnor a significant change from day to day The coefficient of variation was 20 %.

The creatinine excretion rates of smokers and non-smokers in Study B isshown in Table 4.

Smokers had 34 % lower pre-exposure rates than non-smokers, a differencewhich disappeared in the toluene exposed groups but not in the group exposedto clean air This interaction cannot be explained by any of the personal vari-ables.

290

Toluene metabolism during exposure to varying concentrations

Table 4 Urinary creatinine excretion in smokers and non-smokers before and during twoperiods of 7-h exposure to either clean air or 100 ppm toluene (* p< O 05, ** p< O 01).( 1 pmol creatinine = 0 113 mg)

Clean air 100 ppm toluene

14 Non-smokers 10 Smokers 20 Non-smokers 18 SmokersPmol/min g Lmol/min gimol/min ltmol/min

No of subjects Mean SD Mean SD Mean SD Mean SD

Before exposure 11 1 5 3 6 0 5 5 * 8 2 2 8 6 0 3 1 *0-4 h 10 9 4 6 5 3 4 8 ** 8 4 3 0 7 4 3 34-7 h 11 6 3 9 6 4 4 1 ** 9 6 5 0 9 1 5 5

Besides this, a stable correlation to body weight was found in all groups Theregression line of the pre-exposure values is

CER = -1 86 + 0 146 x BW (r = 0 41 P< 0 001)

where CER is creatinine excretion rate in pmol/min ( 1 pmol = 0 113 mg), andBW is body weight in kg.

Females had lower values than males due to the difference in body weight.

Discussion

We have tried to compare the uptake and metabolism of toluene during a con-trolled fluctuating exposure with that of a constant exposure The varying con-centration, which was within the allowed limits of short-term exposure (ACGIH1984-85), gave rise to a considerable variation in alveolar air toluene concentra-tions, reflecting parallel variation in arterial blood toluene levels (Carlsson1982) Moderate physical exercise increased alveolar concentrations by 47 % asin previous studies (Veulemans and Masschelein 1978 ; Carlsson 1982) Theaverage alveolar concentration and the excretion of toluene metabolites werenot changed by the exposure schedule.

In the studies, exercise was performed during the peaks of the varying expo-sure, whereas the subjects rested when the concentration was low This mightimply that the effect of exercise was larger during exposure However, no inter-action between exposure schedule and work level was seen in the excretion ofany of the metabolites, indicating that the effect is small in comparison with thesensitivity of metabolite excretion as an estimate of solvent uptake.

The effect of exercise on HA excretion in Study A is surprisingly small Atthe end of exposure, the rate was higher but the elimination was faster after ex-posure, although the physical activity after exposure on all the days was thesame This paradoxical effect is in accordance with Andersson et al ( 1983),who found a negative correlation between work load and HA excretion after 2-hexposure, and no correlation between uptake of toluene and HA excretion wasseen Veulemans and Masschelein ( 1979) found that exercise increased HAexcretion as expected from data on the toluene uptake The reason for the dif-ferent excretion profile of HA during work may be a redistribution of toluene

291

due to a larger vascularisation of adipose tissue and the muscles (Astrand 1977 ;Billow 1982).

Ortho-cresol excretion was delayed compared with HA and the level wassignificantly increased by physical activity, without changing the time course.Accordingly O-cr seems to reflect toluene exposure over a more extendedperiod than HA, which in the present and in previous studies seems to be re-lated to exposure immediately preceding the urinary sampling (Veulemans andMasschelein 1978 ; Andersson et al 1983).

Standards for HA excretion at the end of an 8-h work shift corresponding toa TLV of 100 ppm have been proposed by a WHO study group ( 1981) and byACGIH ( 1984) The proposed limit of 2 5 g HA/g creatinine ( 1 58 mol HA/molcr) was exceeded by 84 % of the subjects in Study B, and the proposedexcretion rate limit of 3 0 mg HA/min ( 16 8 jpmol/min) was exceeded by 74 %.Thus it seems as if the biological exposure indices are more restrictive than thelimits of work room air.

The large interindividual variation in metabolite excretion is partly due todifferent work level of the subjects, but work level was adjusted to a fixed per-centage of the individual maximal aerobic capacity, and most of the variation isstill unexplained Furthermore, ingestion of alcohol in socially accepted quan-tities inhibits HA and O-cr excretion by 67 and 52 %, respectively, while alveo-lar toluene concentration at the same time is elevated by 40 to 70 % (Waldronet al 1983 ; D O ssing et al 1984).

In a field study Waldron et al ( 1983) found a lower venous toluene concen-tration in habitual drinkers than in non-drinkers, suggesting a faster toluenemetabolism in the former group, and Sato et al ( 1984) showed in short-termanimal studies that ethanol ingested in the hours before exposure to toluene orother solvents increased the metabolism of the solvents.

In this present and in our previous study (D O ssing et al 1983), neither intakeof alcohol the day before exposure nor the daily average consumption influencedtoluene metabolism, possibly because the acute inhibition by alcohol is com-petitive and subsides as soon as alcohol is eliminated from the body Further-more, subjects with a daily consumption of alcohol above 36 g pure ethanolwere excluded, and the intake the day before exposure was generally very low.

The use of creatinine concentration as standardisation of metabolite concen-tration has an advantage compared to estimation of excretion rates as it onlyrequires spot samples However, the sampling period is not well defined, andthere is a considerable interindividual variation, partly related to body weightand sex Moreover, Allesio et al ( 1985) showed a considerable day-to-day var-iation in creatinine excretion rate and the influence of smoking has to be resol-ved The decrease seen in smokers is possibly caused by renal vasoconstrictiondue to the nicotine in the cigarettes as shown in animal experiments (Downeyet al 1982).

Comparison of HA and O-cr as exposure indices shows that HA has a rela-tively low specificity due to the high and variable background level, and thespecificity is expected to decrease further at the actual Danish TLV of 75 ppm.However, HA is easy to measure by relatively cheap and simple analyses (An-dersson et al 1983 ; Hasegawa et al 1983), while the analyses of O-cr are more

292 J Baflum et al.

Toluene metabolism during exposure to varying concentrations

complicated, requiring liquid or gas chromatography preceded by several ex-traction procedures (Pfiffli et al 1979 ; Hansen et al 1982) The percentage ofO-cr as a fraction of HA is rather stable and because of its higher specificity itwould be suitable for work place use instead of HA, especially if the exposurelimit is lowered.

In field studies, HA and O-cr excretion have been correlated to time-weigh-ted average air concentration (TWA) between 0 and 207 ppm (Apostoli et al.1982), between 0 and 125 ppm (Hasegawa et al 1983), and between 10 and61 ppm (De Rosa et al 1985) Excretion of both metabolites were correlatedwith the TWA during the work shift Correlation coefficients between HA con-centration at the end of the shift and TWA were 0 75, 0 82 and 0 80 in the threestudies, while the corresponding coefficient between O-cr and TWA were 0 90,0.61 and 0 68 Hasegawa et al ( 1983) found sex difference in excretion rate andconcentration of HA and O-cr corresponding to the results of our study Mostauthors prefer HA as an index of toluene exposure, but the individual variationin the excretion of both metabolites is large, and only the average of a consider-able number of measurements is a reliable estimate of exposure Moreover,Droz et al ( 1984) used a personal sampler with a flow rate proportional to theactual respiration volume in order to get a better estimate of the individual up-take of toluene This, however, did not show a better correlation to the excre-tion of HA or O-cr than the 8-h TWA Most of the variation, therefore, iscaused by factors other than the individual work rate, which has been one of themajor reasons for preferring biological instead of environmental monitoring ofsolvent.

In conclusion, both HA and O-cr are biological indices of toluene exposure.The interindividual variation of both metabolites, however, is large, and the useof HA and O-cr as biological exposure indices can only serve as a rough esti-mate of the body burden and only on a group basis It is not possible to predictthe individual uptake of toluene by the use of urinary excretion of either HA orO-cr.

Acknowledgements The studies were supported by the Working Environmental Fund, Den-mark and by the Danish Medical Research Fund.

References

Allessio L, Berlin A, Dell Orto A, Toffoletto F, Ghezzi I ( 1985) Reliability of urinarycreatinine as a parameter used to adjust values of urinary biological indicators Int ArchOccup Environ Health 55:99-106

American Conference of Governmental Industrial Hygienists ( 1984) Supplemental documen-tation Biological exposure indices, Cincinnati, Ohio

American Conference of Governmental Industrial Hygienist ( 1984-5) Threshold limit valuesfor chemical substances and physical agents in the work environment Cincinnati, Ohio

Andersen I, Lundqvist GR, M 01 lhave L, Pedersen OF, Proctor DF, Vth M, Wyon DP ( 1983)Human response to controlled levels of toluene in six-hour exposure Scand J Work Envi-ron Health 9:405-418

Andersson R, Carlsson A, Byfalt-Nordqvist M, Sollenberg J ( 1983) Urinary excretion of hip-puric acid and O-cresol after laboratory exposure of humans to toluene Int Arch OccupEnviron Health 53:101-108

293

J Baelum et al.

Apostoli P, Brugnone F, Perbellini L, Cocheo V, Bellomo ML, Silvestri R ( 1982) Bio-monitoring of occupational toluene exposure Int Arch Occup Environ Health 50:153-168

Armitage P ( 1971) Statistical methods in medical research Blackwell Scientific Publications,Oxford

Strand I, Ehrner-Samuel H, Kilbom A, Ovrum P ( 1972) Toluene exposure I Concentrationin alveolar air and blood at rest and during exercise Work Environ Health 9:119-130

Strand PO ( 1977) Textbook of work physiology, 2nd ed McGraw Hill Book Company, NewYork

Bonsnes RW, Tanssky HH ( 1945) On colorimetric determination of creatinine by Jaffe reac-tion J Biol Chem 158:581-591

Bilow J ( 1982) Adipose tissue blood flow during exercise Laegeforeningens forlag, Copen-hagen

Carlsson A ( 1982) Exposure to toluene Uptake, distribution and elimination in man ScandJ Work Environ Health 8:43-55

De Rosa E, Brugnone F, Bartolucci GB, Perbellini L, Bellomo ML, Gori GP, Signon M,Chiesura Corona P ( 1985) The validity of urinary metabolites as indicators of low expo-sures to toluene Int Arch Occup Environ Health 56:135-145

D O ssing M, Baelum J, Hansen SH, Lundqvist GR, Andersen NT ( 1983) Urinary hippuric acidand ortho-cresol excretion in man during experimental exposure to toluene Br J IndHealth 40: 470-473

D O ssing M, Balum J, Hansen SH, Lundqvist GR ( 1984) Effect of ethanol, cimetidine andpropranolol on toluene metabolism in man Int Arch Occup Environ Health 54:309-315

Downey HF, Crystal GJ, Bashour FA ( 1982) Regional renal and splancnic blood flows duringnicotine infusion: effects of alpha and of combined alpha and beta adrenergic blocade JPharmacol Exp Ther 220:375-381

Droz PO, Berode M, Boillat MA, Lob M ( 1984) Biological monitoring and health surveil-lance of rotagravure printing workers exposed to toluene Presented at the ACS meeting,St Louis

Hansen SH, D O ssing M ( 1982) Determination of urinary hippuric acid and O-cresol as indicesof toluene exposure by liquid chromatography on dynamically modified silica J Chro-matogr 229: 141-148

Hasegawa K, Shiojima S, Koizumi A, Ikeda M ( 1983) Hippuric acid and O-cresol in the urineof workers exposed to toluene Int Arch Occup Environ Health 52:197-208

Kjaergaard S, Molhave L, Belum J ( 1985) Variations in the burden of solvent vapours amongprinters Arbejdsmilj O fondet, Copenhagen (In Danish)

Numeric Algorithms Group ( 1980) GENSTAT A general statistic program Rothamsted Ex-perimental Station, UK

Ovrum P, Hultengren M, Lindqvist T ( 1978) Exposure to toluene in a photogravure printingplant Scand J Work Environ Health 4:237-245

Pfaiffli P, Savolainen H, Kalliomaki PL, Kaaliokoski P ( 1979) Urinary O-cresol in toluene ex-posure Scand J Work Environ Health 5:286-289

Sato A, Nakajima T ( 1984) Dietary carbohydrate and ethanol induces alteration of the metab-olism and toxicity of chemical substances Nutr Cancer 6: 121-132

Veulemans H, Masschelein R ( 1978) Experimental human exposure to toluene II Toluenein venous blood during and after exposure Int Arch Environ Health 42:105-117

Veulemans H, Masschelein R ( 1979) Experimental human exposure to toluene III Urinaryhippuric acid excretion as a measure of individual solvent uptake Int Arch EnvironHealth 43: 53-62

Waldron HA, Cherry N, Johnston JD ( 1983) The effects of ethanol on blood toluene concen-tration Int Arch Occup Environ Health 51:365-369

WHO, Technical Report Series no 664 ( 1981) Recommended health-based limits in occupa-tional exposure to selected organic solvents WHO, Geneva

Received October 29, 1986 / Accepted January 9, 1987

294