American Journal of Industrial Medicine 15: 103-110 (1989)

Relationship Between Vapor Exposure and Urinary Metabolite Excretion Among Workers Exposed to Tric h loroet hy lene

Osamu Inoue, MS, Kazunori Seiji, Ms, Toshio Kawai, PhD, Chui Jin, MD, Yu-Tang Liu, MD, Zhen Chen, MD, Shi-Xiong Cai, MD, PhD, Song-Nian Yin, MD, Gui-Lan Li, MD, Haruo Nakatsuka, PhD, Takao Watanabe, PhD, and Masayuki Ikeda, MD, PhD

The exposure-excretion relationship was investigated in 140 trichloroethylene (TR1)- exposed workers and 1 14 nonexposed controls. The time-weighted average intensity of exposure to TRI during the shift as measured by the diffusive sampling method was compared with metabolite levels in the urine collected at the end of the shift in the second half of a working week, when the urinary metabolite levels are expected to reach a maximum. The TRI levels in breathing zone air of the exposed workers were mostly below 50 ppm. The urinary metabolite levels (i.e., total trichloro-compounds, trichlo- roethanol, and trichloroacetic acid) increased as a linear function of the TRI exposure. The relationship between the two exposure indicators was statistically significant in men, women, and both combined. The cross-sectional balance study at the end of the shift revealed that about 4% of TRI absorbed will be excreted at the end of the shift, in agreement with the long biological half-life of this chlorinated hydrocarbon solvent. A possible ethnic difference in the metabolism of TRI is discussed.

Key words: diffusive sampling, ethnic difference, total trichloro-compounds, trichloroacetic acid, trichloroethanol

INTRODUCTION

Biological monitoring is a well-established technique for the estimation of human exposure and is applicable to various industrial chemicals [Alessio et al., 1983, 1986; American Conference of Governmental Industrial Hygienists, 1987;

Department of Environmental Health, Tohoku University School of Medicine, Sendai 980, Japan (0. I . , K.S., H.N., T.W., M.I.). Tohoku Rosai Hospital, Sendai 980, Japan (O.I., K.S.). Kinki Regiondl Safety and Health Service Center, Japan Industrial Safety and Health Association, Osaka 541, Japan (T.K.). Institute of Occupational Medicine, Chinese Academy of Preventive Medicine, Beijing, China (C. J.,

Address reprint requests to Prof. M. Ikeda, Department of Environmental Health, Tohoku University School of Medicine, Sendai 980, Japan. Accepted for publication July 13, 1988.

Y:T.L., J.C., S.-X.C., S.-N.Y., G.-L.L.).

0 1989 Alan R. Liss, Inc.

104 Inoue et al.

Deutsche Forschungsgemeinschaft, 19871, including organic solvents such as aro- matic and chlorinated hydrocarbons.

Although the use of trichloroethylene (TRI) has decreased after its carcinoge- nicity in mice [National Cancer Institute, 19761 was detected, this solvent is still widely utilized for degreasing and should be considered one of the most important chlorinated hydrocarbon solvents used in industry. Statistics indicate that about 73 kilotons of this solvent were sold in Japan in 1985 [Ministry of International Trade and Industries, 19861 and 80.7 kilotons in the United States in 1986 [Storck, 19871; the amounts for tetrachloroethylene and 1 , 1 , 1 -trichloroethane were 72 and 120 kilotons, respectively, in Japan [Ministry of International Trade and Industries, 19861 and 159 and 257 kilotons in the United States [Storck, 19871.

The dose-excretion relationship of TRI has been studied for many years. The precise measurement of the time-weighted average intensity of workplace exposure had been difficult to determine until the diffusive sampling technique was established in recent years [Berlin et al., 19871. A balance study in controlled human exposures has had only limited value because of the long biological half-life (ca. 40 hr [Ikeda and Imamura, 19731) of the solvent.

In the present study, the diffusive sampling technique was applied for the measurement of exposure to TRI during one work shift, and these measures are compared with the TRI metabolite levels in urine. Related reports on the metabolism of aromatic solvents in humans have been published elsewhere [Hasegawa et al., 1983; Inoue et al., 1986a,b, 1988a,b].

MATERIALS AND METHODS Population

The survey was conducted in the second half of a working week, when the urinary metabolite levels were expected to be maximum [Ikeda and Hara, 19801. The study population, 140 exposed and 114 nonexposed workers, were employed in two factories in northeast China. Plant A produced TRI by chlorination of acetylene followed by dehydrochlorination. In plant B, a metal parts plant, TRI was used for degreasing prior to metal plating. In both plants, the exposed workers (61 men and 17 women from plant A and 52 men and 10 women from plant B) served on three shifts [0800-1600, 1600-0000 (midnight) and 0000-08001, while the controls worked 0800-1600 in the mechanical sections of the plants (52 men and 22 women from plant A and 36 men and 4 women from plant B).

Exposure Measurement

Measurements were made using personal diffusive samplers with carbon cloth K-filter 1500 (Toyobo Co., Osaka, Japan) as absorbent [Hirayama and Ikeda, 1979; Ikeda et al., 1984; Kasahara and Ikeda, 19871. After the exposure, TRI in the carbon cloth was extracted with carbon disulfide (spiked with sec.-butylbenzene as an internal standard [Kasahara and Ikeda, 1987]), and the extract was analyzed with an automated capillary GC system, which consisted of a liquid autosampler (Hitachi model 163-08), a FID-GC (Hitachi model 163) equipped with a 50-m-long Silicone OV 101 FS-WCOT column [Kasahara et al., 19871, and a chromato-integrator (Hitachi model D-2000). TRI vapors of known concentrations for calibration were

Exposure-Excretion in TRI-exposed Workers 105

TABLE 1. Trichloroethylene (TR1)-exposed Workers in Two Plants

Men Women

Plant No." T R I ~ TTC' NO.^ T R I ~ TTC' ~

A 61 3 -94 5 127 17 2-47 5 111 B 52 1-63 5 89 10 2-1 3 5 98

Plant A was a TRI-synthesizing plant, and plant B was a plant in which TRI was used for degreasing. "Nos. of exposed workers.

'Urinary total trichloro-compounds (TTC) level observed (mg/L). Range of TRI exposure (ppm) in personal samples.

generated by means of a servomechanized vapor generation system [Koizumi and Ikeda, 1981; Kumai et al., 19841.

U ri na I yses Workers urinated about 2 hr before the end of the shift, and urine samples were

then collected at the end of the shift. Urinary total trichloro-compounds (TTC), trichloroethanol (TCE), and trichloroacetic acid (TCA) were measured according to Tanaka and Ikeda [1968]. The detection limit was 0.5 mg/L. The results were expressed either as observed or after correction for creatinine concentration [Jackson, 19661 or specific gravity of urine of 1.016 [Rainsford and Lloyd Davies, 19651.

Statistical Analysis Regression analysis and student's t-test were conducted utilizing a package

program with a personal computer (PC-9801VM21, NEC Corporation, Tokyo, Japan).

RESULTS AND DISCUSSION

In total, 140 exposed workers and 114 nonexposed controls were examined. Contrary to the general assumption that the concentration would be higher in degreasing workshops than in TRI synthesis plants, the intensity of exposure to TRI appeared to be somewhat higher in plant A than in plant B, as the maximum concentrations suggest (Table I). The maximum TTC levels were also higher in plant A than in plant B. The urinary metabolite levels of the controls were essentially zero (i.e., below the detection limit of 0.5 mg/l).

The urinary metabolite levels (TTC, TCE, and TCA) were further corrected for creatinine concentration [Jackson, 19661 as well as a specific gravity of 1.016 [Rainsford and Lloyd Davies, 19651, The correlation was examined by regression analysis between the time-weighted average intensity of exposure and the urinary metabolite levels either noncorrected (observed values) or corrected by the two methods. Because the regression lines obtained with the plant A data were quite similar to the counterpart lines with plant B data, the data from two plants were pooled and subjected to the analysis by metabolite (i.e., for TTC, TCE, and TCA, respectively) and by sex (i.e., for men and women). The TTC and TCE levels were significantly (p < .05 and p < .01, respectively) higher in men than in women, while TCA levels did not differ significantly (p > . lo) between the two sexes. For practical reasons, the calculation was then made for the two sexes combined. The results are

106 Inoue et al.

summarized in Table 11. The relationship of the urinary metabolite levels (observed values) with the breathing zone TRI concentrations are depicted as scatter diagrams in Figure 1, in which the calculated regression lines and the 95% confidence ranges for the group means and individual values are also shown.

It appears that the linear relation holds for all the cases examined (Table 11). The correlation coefficients were all statistically significant (p < .01 in all cases), although the value tended to be smaller with TCA than with TTC, especially among men. The values for TTC were less than .7 at best, in contrast to the cases of phenol excretion after benzene exposure in which the correlation coefficients were better than .8 in general [Inoue et al., 1986al. The smaller correlation coefficient with TRI is not unexpected; after exposure to TRI with a biological half-life of about 40 hr [Ikeda and Imamura, 19731, the metabolite levels will be affected by the exposure not only on the day of study but also on the preceding days; such is less likely for benzene with a half-life of less than 10 hr [Ikeda and Hara, 19801.

Apparent saturation of TCA excretion was observed at the TRI concentration of 50-100 ppm in a previous study [Ikeda et al., 19721. It was also reported that the urinary excretion of TTC will level off at high doses of TRI, for example, after repeated oral administrations of 1,600 mg TRUkg/day for 6 months to rats [Buben and O’Flaherty, 19851 or after a single intraperitoneal injection of 1,460-2,820 mg TRUkg to mice [Rouisse and Chakrabarti, 19861. Findings to suggest saturation of metabolic capacity were not obtained in the present study, possibly because the exposure was much less intense and well below 50 ppm for most of the workers examined. Thus, it is plausible to deduce that the capacity of humans to metabolize TRI will not be exceeded at the current occupational exposure limit of 50 pprn [Japan Association of Industrial Health, 1987; American Conference of Governmental Industrial Hygienists, 1987; Deutsche Forschungsgemeinschaft, 19871,

While follow-up of urinary metabolite excretion could not be made in the present study, it is possible to calculate a cross-sectional balance between the amount of TRI absorbed through the lungs and the excretion of metabolites into urine at the end of a shift with TRI exposure of, for example, 10 ppm (= 53.8 mg/m3), utilizing the parameters obtained in the present study (Table 11). For calculation, it was assumed that about 50% inhaled TRI will be absorbed (e.g., 35% according to Nomiyama and Nomiyama [1971] and 63% according to Stewart et al. [1970]) and that the rates of respiration and urine excretion are 15 Umin and 1 ml/min, respec- tively. The amount absorbed will be 53.8 (mg/m3) X 15 (Umin) X lop3 (m3/l) X 0.5 = 0.403 (mg/min), and the amount of TTC (as TCA) to be excreted into urine will be 21.47 (mg/l) X 131.4/163.4 X 1 (ml/min) X = 0.017 (mg/min), where 21.47 (mg/l) is obtained from the regression equation for TTC (observed value for the combination of men and women; Table 11) as 7.87 + (1.36 X lo), and 13 1.4 and 163.4 are the molecular weights of TRI and TCA, respectively.

The rate of the excreted amount over the absorbed amount, that is, 0.017/0.403 or 4.2%, is even lower than the value (ca. 20%, as recalculated with a respiration rate of 15 l/min) previously estimated for Japanese [Ikeda et al., 19721. This low rate is in agreement with the fact that the biological half-life of TRI measured by urinary metabolite levels after the cessation of exposure is as long as 40 hr. This is exceeded only by the half-life of tetrachloroethylene (140 hr) among commonly used solvents [Ikeda and Imamura, 1973; Ikeda and Hara, 19801, as previously discussed [Ikeda et al., 19721.

TA

BL

E 1

1. C

alcu

late

d R

egre

ssio

n L

ines

by

Met

abol

ite a

nd b

y Se

x in

TC

E-e

xpos

ed w

orke

rs

Para

met

er o

f th

e re

gres

sion

lin

e" w

ith th

e m

etab

olite

lev

el a

s

Cor

rect

ed f

or

Obs

erve

d'

spec

ific

grav

ityd

Cor

rect

ed f

or c

reat

inin

e' N

o.

Met

abol

ite

Sex

wor

kers

b a

P rf

a P

If

U

P rf

Tota

l tri

chlo

ro-

com

poun

ds (

TTC

) M

en

20 1

W

omen

53

C

ombi

ned

254

Tric

hlor

oeth

anal

Men

20

1

Wom

en

53

Com

bine

d 25

4

Tric

hlor

oecc

tic

acid

(TC

A)

Men

20

1

Wom

en

53

Com

bine

d 25

4

8.36

5.

80

7.87

(2

.50)

5.90

2.

74

5.19

(1

.95)

2.82

3.

06

2.97

(1

.13)

1.36

1.

42

I .36

(0

.19)

1.03

0.

81

0.99

(0

.15)

0.31

0.

61

0.36

(0

.09)

.677

,5

89

,657

.648

,5

65

,629

,438

,5

47

.457

6.57

3.

98

6.06

(1

.99)

4.15

1.

91

3.63

(1

.36)

2.04

2.

08

2.12

(0

.81)

0.99

1 .o

o 0.

99

(0.1

5)

0.74

0.

55

0.70

(0

.11)

0.23

0.

44

0.27

(0

.06)

,622

,6

22

,620

,655

.5

80

,636

,440

,5

93

,470

7.42

1.

07

.605

3.

52

1.27

.6

96

6.71

1.

10

.619

(2

.23)

(1

.73)

4.88

0.

83

.611

1.

66

0.65

,6

37

4.17

0.

79

.605

(1

.66)

(0

.13)

2.60

0.

25

,328

1.

84

0.63

,6

52

2.58

0.

31

,394

(1

.16)

(0

.09)

,

,;

,

.,

,_

Nos

. in

the

tab

le a

re t

he p

aram

eter

s as

indi

cate

d. T

hose

in

pare

nthe

ses

are

the

valu

es t

o ca

lcul

ate

the

95%

rang

es o

f th

e co

rres

pond

ing

para

met

ers

for

men

and

wom

en c

ombi

ned,

for

exa

mpl

e, 7

.87

(2.5

0) m

eans

tha

t the

95%

rang

e fo

r th

e pa

ram

eter

s of

the

reg

ress

ion

line

is

act

and

P ar

e th

e in

terc

ept a

nd th

e sl

ope

of t

he c

alcu

late

d re

gres

sion

lin

e so

that

Y =

u +

PX, w

here

X i

s th

e T

RI

conc

entra

tion

(in p

pm)

in b

reat

hing

zon

e ai

r an

d Y

is th

e co

ncen

tratio

n of

TT

C, T

CE

, or T

CA

in

urin

e.

bExp

osed

and

con

trol

wor

kers

com

bine

d.

'Met

abol

ite c

once

ntra

tion

as o

bser

ved

(uni

t; m

g T

CN

L).

dM

etab

olite

con

cent

ratio

n ad

just

ed f

or s

peci

fic g

ravi

ty o

f ur

ine

of 1

.016

(uni

t; m

g T

CM

L).

'M

etab

olite

Con

cent

ratio

n di

vide

d by

cre

atin

ine

conc

entra

tion

(uni

t; m

g T

CA

/g c

reat

inin

e).

r in

dica

tes

the

corr

elat

ion

coef

ficie

nt;

all o

f th

em a

re s

tatis

tical

ly s

igni

fican

t (p

< .0

1).

5.37

(=

7.8

7 - 2

.50)

to 1

0.37

(=

7.8

7 +

2.50

).

f.

108 Inoue et al.

150

100

50

0

B. TRICHLOROETlIANOL

C. TRICHLOROACETIC ACID 1 _---

0 2 0 4 0 60 80 100

TRICHLOROETHYLENE I N A I R (pp in) Fig. 1. Relationship between TRI concentrations in the breathing zone air of workers and metabolite concentrations (observed, noncorrected values) in the urine. A, TTC; B, TCE; C, TCA. Points indicate individual values. Lines and curves are calculated regression line (solid line in the center), 95% confidence ranges of sample means (curves on both sides of the regression line), and 95% confidence ranges of the individual samples (the outmost lines). Nonexposed values are not depicted to avoid congestion on the vertical axis so that the intercept on the axis is clearly seen.

Exposure-Excretion in TRI-exposed Workers 109

Detection tube data were employed for the Japanese study [Ikeda et al., 19721 because the diffusive sampling technique was not yet available, whereas the diffusive sampling method was utilized in the present study on Chinese workers for more precise determination of exposure intensity. Such difference in the method of exposure measurement may explain the variation in the findings, at least in part. Alternatively, it may also be possible to speculate that the observed difference in the metabolic rates (i.e., 20% for Japanese and 4% for Chinese) is due to ethnic difference in the subjects studied, as it is known that Chinese workers excrete less hippuric acid than do the Japanese after occupational exposure to toluene at compa- rable levels [Inoue et al., 1986bl. The latter possibility apparently deserves further study to elucidate if a difference in solvent metabolism exists, in general, among populations of various ethnic origins.

CONCLUSIONS

In the present study, the relationship between the time-weighted average intensity of exposure to trichloroethylene (TRI) during a work shift and the metabolite levels in urine collected at the end of the shift was investigated in 140 exposed workers and 114 control workers in the second half of a working week. The intensity of exposure was less than 50 ppm in most cases studied. There is a linear correlation between TRI levels in breathing zone air and urinary levels of each of the three metabolites (i .e., total trichloro-compounds, trichloroethanol, and trichloroacetic acid) in men, women, and both combined. A balance study at the end of the shift suggests that about 4% of TRI absorbed via lungs will be excreted in urine.

ACKNOWLEDGMENTS

The authors are grateful to Drs. Feng-Lin Zhu, Xiang-Rong Li, and Iun-Ming Lei (Shenyang, China) and Drs. Shu-Mei Cheng, Ting-Qin Yang, Chu-Xia Ren, and Hai-Cheng Yu (Jinxi, China) for their support and collaboration in the factory survey. Thanks are also due to Prof. T. Suzuki (Tohoku Rosai Hospital, Sendai, Japan) and Prof. M. Tati (Occupational Health Service Center, Japan Industrial Safety and Health Association, Tokyo, Japan) for their interest in this work.

REFERENCES

Alessio L, Berlin A, Roi R, Boni M (eds.) (1983): “Human Biological Monitoring of Industrial Chemicals Series.” Ispra: Joint Research Centre Ispra Establishment.

Alessio L, Berlin A, Roi R, Boni M (eds.) (1986): “Biological Indicators for the Assessment of Human Exposure to Industrial Chemicals.” Ispra: Joint Research Centre Ispra Establishment.

American Conference of Governmental Industrial Hygienists. (1987): “Threshold Limit Values and Biological Exposure Indices for 1987-1988.” Cincinnati: ACGIH.

Berlin A, Brown RH, Salinders KJ (eds.) (1987): “Diffusive Sampling-An Alternative Approach to Workplace Air Monitoring.” London: Royal Society of Chemistry.

Buben JA, O’Flaherty EJ (1985): Delineation of the role of metabolism in the hepatotoxicity of trichloroethylene and perchloroethylene: A dose-effect study. Toxic01 Appl Pharmacol 78: 105-122.

Deutsche Forschungsgemeinschaft (1987): “Maximale Arbeitsplatzkonzentrationen und Biologische Arbeitsstofftoleranzwerte 1987. ” Weinheim: Verlag Chemie (in German).

110 Inoue et al.

Hasegawa K, Shiojima S, Koizumi A, Ikeda M (1983): Hippuric acid and 0-cresol in the urine of workers exposed to toluene. Int Arch Occup Environ Health 52:197-208.

Hirayama T, Ikeda M (1979): Applicability of carbon felt to the dosimetry of solvent vapor mixture. Am Ind Hyg Assoc J 40:1091-1096.

Ikeda M, Hara I (1980): Evaluation of the exposure to organic solvents by means of urinalysis for metabolites. Sangyo Igaku (Jpn J Ind Health) 22:3-17 (in Japanese).

Ikeda M, Imamura T (1973): Biological half-time of trichloroethylene and tetrachloroethylene in human subjects. Int Arch Occup Environ Health 31:209-224.

Ikeda M, Ohtsuji H, Imamura T, Komoike Y (1972): Urinary excretion of total trichloro-compounds, trichloroethanol and trichloroacetic acid as a measure of exposure to trichloroethylene and tetrachloroethylene. Br J Ind Med 29:328-333.

Ikeda M, Kumai M, Aksoy M (1984): Application of carbon felt dosimetry to field studies distant from analytical laboratory. Ind Health 22:53-58.

Inoue 0, Seiji K, Kasahara M, Nakatsuka H, Watanabe T, Yin S-N, Li G-L, Jin C, Cai S-X, Wang X-Z, Ikeda M (1986a): Quantitative relation of urinary phenol levels to breathzone benzene concentra- tions; A factory survey. Br J Ind Med 43:692-697.

Inoue 0, Seiji K, Watanabe T, Kasahara M, Nakatsuka H, Yin S-N, Li G-L, Cai S-X, Jin C, Ikeda M (1986b): Possible ethnic difference in toluene metabolism: A comparative study among Chinese, Turkish and Japanese solvent workers. Toxicol Lett 34: 167-174.

Inoue 0, Seiji K , Watanabe T, Kasahara M, Nakatsuka H, Yin S-N, Li G-L, Cai S-X, Jin C, Ikeda M (1988a): Mutual metabolic suppression between benzene and toluene in man. Int Arch Occup Environ Health, 60:15-20.

Inoue 0, Seiji K, Watanabe T, Kasahara M, Nakatsuka H, Yin S-N, Li G-L, Cai S-X, Jin C, Ikeda M (1988b): Determination of catechol and quinol in the urine of workers exposed to benzene. Br J Ind Med, 45:487-492.

Jackson S (1966): Creatinine in urine as an index of urinary excretion rate. Health Phys 12:843-850. Japan Association of Industrial Health (1987): Recommended occupational exposure limits for 1987.

Kasahara M, Ikeda M (1987): Spontaneous desorption of organic solvents from carbon cloth. Ind Health

Kasahara M, Suzuki H, Takeuchi Y, Hara I, Ikeda M (1987): n-Hexane, benzene and other aromatic components in petroleum distillate solvents in Japan. Ind Health 25:205-214.

Koizumi A, Ikeda M (I98 1): A servomechanism for vapor concentration control in experimental exposure chambers. Am Ind Hyg Assoc J 42:417-425.

Kumai M, Koizumi A, Morita K, Ikeda M (1984): An exposure system for organic solvent vapor. Bull Environ Contam Toxicol 32:200-204.

Ministry of International Trade and Industry (Research and Statistics Department, Minister’s Secretariat), the Government of Japan (1985): “Yearbook of Chemical Industries Statistics 1985.” Tokyo: Tsusho Sangyo Chosa Kai, p. 88.

National Cancer Institute (1976): “Carcinogenesis Bioassay of Trichloroethylene. National Cancer Institute Carcinogenesis Technical Report Series No. 2, NCI-CG-TR-2.” Bethesda, MD: National Institutes of Health.

Nomiyama K, Nomiyama H (1971): Metabolism of trichloroethylene in humans; Sex difference in urinary excretion of trichloroacetic acid and trichloroethanol. Int Arch Arbeitsmed 28:37-48.

Rainsford SG, Lloyd Davies TA (1965): Urinary excretion of phenol by men exposed to vapour of benzene: A screening test. Br J Ind Med 22:21-26.

Rouisse L, Chakrabarti SK (1986): Dose-dependent metabolism of trichloroethylene and its relevance to hepatotoxicity in rats. Environ Res 40:450-458.

Stewart RD, Dodd HC, Gay HH, Erley DS (1970): Experimental human exposure to trichloroethylene. Arch Environ Health 20:64-71,

Storck W (1987): Chlorinated solvent use hurt by federal rules. Chem Eng News Sept. 28, p. 11. Tanaka S, Ikeda M (1968): A method for determination of trichloroethanol and trichloroacetic acid in

Sang0 Igahn Jpn J Ind Health 29:397-415 (in Japanese).

25173-81.

urine. Br J Ind Med 25:214-219.