Carbon disulphide

© Springer-Verlag 1995

Int Arch Occup Environ Health (1995) 67:5-10 C

H Drexler

I~IAI~II~mnn p

Th G6en · J Angerer

Carbon disulphide II Investigations on the uptake of C5 2 and the excretion of its metabolite 2-thiothiazolidine -carboxylic acid after occupational exposure

Received: 16 June 1994/Accepted: 7 October 1994

Abstract The reported investigations on the uptake of carbon disulphide (C 5 2) and the excretion of its metabolite 2-thiothiazolidine-4-carboxylic acid (TTCA) were based on results from 403 personal air samples (352 passive and 51 active samples) and 362 TTCA determinations in biological material measured during a field study on the adverse effects due to C 5 2 exposure. The external exposure ranged from below the detection limit (0 2 ppm) to 66 ppm and the urinary TTCA excretion from below the detection limit (0 16 mg /l) to 33.4 mg/1 The excretion of TTCA in postshift urine related to creatinine and volume showed a linear correlation to the C 52 air concentration On the basis of these results the influence on the internal exposure of physical work load, dermal exposure and individual parameters (age, Brocaindex, disturbed skin barrier) was evaluated Correlations between the TTCA values in the postshift urine and the individually measured C5 2 concentrations were carried out separately for individual departments and persons with and without indications of a disturbed skin barrier In order to be able to judge the individual internal exposure related to external exposure, a personal quotient was formed from the TTCA level in the urine and the C 5 2 air concentration measured on the same day (relative interal exposure RIE index = TTCA mg/g creatinine/C 52 in ppm) On investigating interindividual differences, higher relative internal exposures were found in persons with a heavy physical work load and more intensive skin contact It could be shown for a large group of persons exposed to C 52 that a pathological skin

Dedicated to Professor G Zehnert on the occassion of his 65th birthday H Drexler (E) · Th G 6en · J Angerer Institute and Out-patient Clinic for Occupational, Social and Environmental Medicine of the University of Erlangen Nuremberg, Schillerstrasse 25/29, D-91054 Erlangen, Germany

condition leads to an increase in the dermal penetration rate of hazardous substances By means of the RIE index it could be shown that the TTCA excretion related to the individual external exposure increases significantly with a decreasing Broca index, which must be taken into consideration with greatly overweight persons and exposures in the range of the currently valid threshold limit values The interindividual differences in internal exposure found at the same ambient air concentration emphasize the importance of biological monitoring for individual health protection and the setting of biological threshold limit values. Key words Carbon disulphide · 2-Thiothiazolidine-4carboxylic acid · Biological monitoring Dermal penetration · Skin disease

Introduction In many cases the updated dose of a hazardous substance and the corresponding health risk to the individual can only be estimated approximately by the determination of the concentration of chemical substances in air This is particularly true for chemicals which can enter the organism via the skin and also have great toxic potential Carbon disulphide (C5 2), an important reagent in the production of viscose, is an example of such substances. Under usual workplace conditions inhalation is a more important route of uptake of C 5 2 compared to absorption via the intact skin (Beauchamp et al 1983 ; WHO 1979 ; BUA 1993) At workplaces with intensive skin contact with C 52 , however, the main route of absorption can be via the skin The horny layer, which is only 10-100 tlm thick, forms a barrier to the absorption of chemical substances Compounds with hydrophilic and hydrophobic molecular parts penetrate this barrier best of all (Berner and Cooper 1987).

6

Measurement of the transepidermal water loss proves that the barrier function of the skin is already damaged in the presence of any dermatosis and skin irritation at sub-clinical stages (Fartasch et al 1992 ; Aalto-Korte and Turpeinen 1993 ; Lavrijsen et al 1993) It can therefore be assumed that every pathological skin condition causes increased percutaneous absorption This is also a reason why the uptake of C 52 can take place predominantly via the skin. With inhalational uptake, after an initial phase an equilibrium is reached after about 2 h and approximately 60-70 % of the inhaled C 5 2 is retained (Beauchamp et al 1983) As the equilibrium is on the side of retention, an increase in the respiratory minute volume causes an increased uptake of C 5 2, so that the effective internal exposure increases with physical activity Therefore, at workplaces with the same ambient air concentration but different physical demands the resulting internal exposure should differ. The C 52 metabolite 2-thiothiazolidine-4-carboxylic acid (TTCA) has proven a specific and sensitive indicator of the internal exposure (Campbell etal 1985; Rosier et al 1982 ; Drexler et al 1994) Half-lives of 2-8 h are quoted for the excretion of TTCA (Rosier et al 1987; Freudlsperger and Madaus 1988) Accumulation therefore seems possible after higher C 5 2 exposures. With regard to the valid quantification of the relevant internal exposure to C 5 2 the following questions seemed important: 1 Which kinetic characteristics of C 52 and its metabolite TTCA should be considered during biological monitoring? 2 Is the relationship between external and internal exposure to C5 2 dependent on the workplace or the kind of working activity? 3 Are there interindividual differences in the relationship of external to internal C5 2 exposure?

Materials and methods The data were collected in a field study on employees of viscoseproducing factory The description of the group can be found in Table 1 Altogether 362 male persons between the age of 21 and 60 were investigated who had worked for up to 44 years at workplaces exposed to C 52 The workers were employed in five work areas separate from each other (spool spinning department (1), spinning department (2), washing department (3), post-treatment of technical rayon (4), ageing of viscose and filtration (5)l A more detailed description of this group can be found in part I of this study (Drexler et al 1994) The time of investigation was selected according to the production schedule Therefore the persons were examined on different days of their shift period About 95 % of the workers were investigated between the 1st and 7th days of the shift. Detailed anamnestic data for an investigation of cardiovascular effects of C52, which are not presented here, including alcohol and nicotine consumption, were collected and the Broca index, lweight in kg/ (height in cm 100)l x 100, was calculated In order to answer the question of whether a pathological skin condition causes

Table 1 Characterisation of the exposed group Median

range

Age (years)

40

21

60

Height (cm)

176

166

203

Weight (kg)

82

57

120

Broca index (%) Total length of exposure (months) C 52, air (ppm) TTCA, urine (mg/g creatinine) a

107 1 108 39 1 63

75 4 < 0 20 < 0 16

166 528 65 7a 11 57

Highest value: 65 7 ppm; second highest value: 28 1 ppm

a higher internal exposure, the test persons were asked about the frequency of skin irritation and current skin diseases were recorded. Permanent skin irritation or the occurrence of it several times a month was regarded as an indication of damage to the skin barrier. Skin diseases documented by a dermatologist present at the time of the investigation and florid skin irritation were described as "current skin disease" Definition of a damaged skin barrier (skin irritation per month and/or florid skin disease) was made before the beginning of the statistical evaluation. On the day of the investigation the personal air exposure to C52 of the employee was measured using passive air samplers For the investigation of TTCA at the end of the shift a spot sample of urine was taken The description of the analytical methods can be found in the first part of this study (Drexler et al 1994). In order to check whether different relationships between external and internal exposure to C 52 could be determined in different subgroups, the correlations between individual C 52 air concentrations and TTCA excretions were investigated separately for the various departments and for persons with expected disturbance of the skin barrier as compared to persons with no conspicuous skin condition. The individual cumulative duration of exposure (in months) was calculated to investigate a possible influence of long-term exposure on the metabolism of C52. The external exposure ranged from below the detection limit (0.2 ppm) to 65 7 ppm We therefore calculated a quotient from the TTCA concentration in the urine of the employee (in mg/g creatinine) and the C 52 air concentration (in ppm): RIE = TTCA (mg/g creatinine)/C 52 (ppm). In the following this quotient is referred to as TTCA/C 52 or as the RIE index We regard this quotient as the measure of the internal exposure related to the individual exposure On this basis we were able to compare the individual relationships of internal to external exposure. Using linear regression analyses we investigated whether these RIE indices were influenced by the Broca index, the day of the shift, age, the total duration of exposure and nicotine and alcohol consumption Eleven persons lost their passive monitors during measurement In the case of one person no creatinine determination in urine could be carried out There are therefore 350 pairs of values. Formation and evaluation of the RIE index only seemed justifiable, however, where both values were above the detection limit This was the case in 332 test persons (Drexler et al 1994). Statistical data analysis was carried out using SPSS for Windows (SPSS Inc ) The difference between the linear regressions was calculated according to the formulas described The null hypothesis tested assumes a parallelism of two curves (Sachs 1992).

Results

Figure 1 shows the distribution of the RIE-indices The RIE indices show that with the same external exposure

7 Number (n)

RIE -Index 1,0-

,8-

I

,6-

L 4-

r rri

-I

_

7 -

Ll~l

<

v

TCNC 52 (R IE index)

I_

j

i

,2-

1

Fig 1 Distribution of the RIE indices (mg TTCA per gram creatinine/C 5 2 in ppm) oO o N=

(ambient air concentration) there is considerable interindividual variation in TTCA excretion The relative internal exposure (RIE index) for the whole group averaged 0 495 mg (g creatinine ppm) '

32

44

48

50

57

41

45

1st

2nd

3rd

4th

5th

6th

7th

day of working period

Fig 2 Boxplot of RIE index for the day of the shift Medians, 50% ranges and 95 % ranges are indicated

The RIE index shows a negative correlation with the Broca index The dependence can be described by: TTCA/C 5 2 =

0 00346' Broca index + 0 860

(P < 0 001).

Table 2 Differences in the RIE indices (in mg TTCA per gram creatinine/C 52 in ppm) of the spinners compared to the workers in the other departments

Broca index (%) = lbody weight (kg)/height (cm) 100 l x 100.

Accordingly, the following relationship to the Broca index variation (A Broca index) applies for the difference of the TTCA excretion (A TTCA): ATTCA =

Mean SD Median 95th percentile Min Max No

Spinners

Others

0 515 0 258 0 461 0 907 0 123 1 732 198

0 464 0 315 0 409 0 958 0 128 3 087 134

0 0034 C 5 2' A Broca index.

At the level of the currently valid MAK value for C 5 2 ( 10 ppm) there is a TTCA difference of 0 346 mg/g creatinine per 10% difference in the Broca index In each case it was first checked that the Broca indices were distributed evenly in the sub-groups Thus, in all further calculations confounding due to the Broca index could be excluded. A dependency of the quotient TTCA/C 5 2 on age, total duration of exposure and alcohol and nicotine consumption could not be found Likewise, the day of the shift had no influence on the RIE index (Fig 2). On considering the RIE indices there was a statistically significant difference between employees in the spinning rooms (departments 1 and 2) and those in the other departments (Table 2) This difference remained when only persons without skin disease and/or skin irritation were considered. In Table 3 the linear regressions between external and internal exposure are represented for the individual subgroups For departments 1, 2 and 4 the equations show almost identical increases The linear regressions for departments 3 and 5 are different However, in both departments the number of cases was small, which also

Table 3 Comparison of the linear regressions in the various subgroups ly = TTCA (mg/g creatinine); x = C 52 (ppm)l Subgroups

TTCA on selection of C52

r

No.

Department 1 Department 2 Department 3 Department 4 Department 5 Healthy skin Diseased skin Healthy-skinned spinners Spinners with diseased skin

y = 0 288 y = 0 346 y = 0 158 y = 0 293 y = 0 176 y = 0 287 y = 0 406 y = 0 347

0 762 0 658 0 752 0 665 0 416 * 0 799 0 789 0 783

109 93 37 95 16 279 71 147

x x x x x x x x

+ + + +

0 47 0 92 1 92 0 45 + 0 49* + 0 65 + 0 32 + 0 49

y = 0 419 x + 0 40

0 694

56

* Not significant (P > 0 05)

seems to be a reason why the correlation for department 5 probably could not be ascertained For department 3 there was, moreover, a great range of scatter for the exposure.

8 Persons exposed to C52

Spinning Department

'12 .

with skin irritation/disease 11

0 10 U

12 C 11 5'10

9

no skin irritation/disease

i,

A

E c k.

C

7 6 5 4

5 r =

9 8 7

with skin irritation/disease no skin irritation/disease

6 5

4

: 3 2

3 -# * i

2 .

0 0 2

LLY

4 6 8 10 12 14 16 182022 242628 30 32 C52-oir (ppm)

Fig 3 a Correlation between C52 concentration in air (personal air sampling) and TTCA excretion (mg/g creatinine) in urine for all persons exposed to C 52 with and without skin irritation or skin disease (0, with skin irritation/disease; A, without skin irritation/disease) b Correlation between C 52 concentration in air (personal air sampling) and TTCA excretion (mg/g creatinine) in urine for persons with and without skin irritation or skin disease in the spinning department (0, with skin irritation/disease; A, without skin irritation/disease)

Persons with frequent skin irritation (continuous or several times per month, N = 61) and those with current skin disease at the time of the investigation (eczema, psoriasis, folliculitis, N = 20) showed a steeper increase in the linear regression than people with healthy skin (Fig 3 a) With a probability of error of < 5 % these are real differences Considering the persons with skin diseases or skin irritations in isolation the same trend is visible This trend can also be seen when only those persons employed in the spinning rooms with and without dermatosis are compared (Fig 3 b).

Discussion We investigated in parallel in a sufficiently large group the personal external C5 2 air concentration and the internal exposure, represented by the C 52 metabolite TTCA Thus we were able to calculate a correlation between external and internal exposure By comparing the correlation in certain groups we found differences due to uptake, excretion and dermal exposure As a measure of the relative internal exposure for exposed workers we formed a quotient (RIE index) from the amount of TTCA in the post-shift urine related to creatinine and the worker's C 5 2 air value as ascertained by personal air sampling Assuming that the scatter of these quotients is not only a product of deviation from the true value, unavoidable during measurement, a high quotient indicates (compared to the ambient exposure) a relatively high internal exposure and a low

1 0

l l

0 2

4 6

l

l

@ l

l

l

l

l

8 10 12 14 16 1820222426283032 C52-air (ppm)

quotient, a relatively low internal exposure The plausible and statistically significant differences we found support this premise The scatter of these quotients suggests that exact quantification of C 5 2 exposure is only possible using biological monitoring This is of particular relevance when observed clinical effects are to be attributed to the dose or when a no observed effect level (NOEL) is to be estimated. The negative correlation we found between the TTCA excretion related to the individual ambient exposure and the Broca index must be taken into consideration when evaluating the results, particularly when there is coincidence of high C 5 2 exposures and very high body weight After external exposure at the level of the currently valid MAK value, an exposed worker who is 40 % overweight would excrete 1 38 mg TTCA/g creatinine less than someone of normal weight (BEI value 5 mg TTCA/g creatinine) We interpret the dependency of the RIE index on the Broca index as an indication that a part of the C 5 2 reaches the adipose tissue as a third compartment and is metabolised either delayed or in some other way There is probably, however, no long-term storage in the adipose tissue with subsequent retransportation into the blood as we found no accumulation in the post-shift urine in the course of a working week. These results agree with those of Van Poucke et al. ( 1990) In the pre-shift urine of persons exposed to C 5 2 they found increasing TTCA concentrations from the 1st to the 3rd day, which possibly can be explained by a temporary storage of C 5 2 in the adipose tissue Again, however, this had no influence on the concentrations in the post-shift urine. Relative internal exposure was not found to be dependent on age, the total length of exposure, the day of the shift or alcohol and nicotine consumption We regard this as an indication that these factors cause no noteworthy deviation in the C52 metabolism due to enzymatic induction. As described in the introduction, the internal exposure to a substance with the same external exposure can

9

vary due to a variety of individual factors Differences in respiratory minute volumes due to differences in physical work load and constitution, as well as additional percutaneous absorption, should be taken into consideration after exposure to C 5 2 as variables affecting the uptake of the substance Probable causes of the higher relative internal exposures of the spinners compared to the other exposed employees are an increased respiratory minute volume and a greater percutaneous absorption rate among the spinners In the spinning rooms up to 18 kg spools have to be moved by hand in a prescribed time The work processes in the other departments are, by contrast, automated to a large extent (Drexler et al 1994) The work load in the spinning rooms therefore differs from that of the other workplaces with exposure to C 52 due to, among other things, the heavy physical work. The C 52 which is necessary to transform the cellulose into soluble xanthogenate is released in the spinning process For this reason the spinning machines are shielded and ventilated Whenever manual work has to be carried out it is unavoidable that the viscose spinner has to work in the area of these machines, usually, according to our observations, with unprotected arms. In the machines there are much higher C5 2 concentrations than in the ambient air Uptake of C 52 via the skin may be assumed to occur both via direct skin contact with the spinning solution containing C 5 2 and directly from the air In animal experiments the dermal absorption of C 5 2 from the vapour phase has been proven (Cohen et al 1958). This percutaneous absorption of C 52 is even possible for intact skin in toxicologically significant amounts (Dutkiewicz and Baranowska 1967) With damage to the epidermal barrier as a result of a pathological skin condition an increase in the percutaneous penetration rate is to be expected This hypothesis has also been put forward by Grandjean (1990) However, he was able only to refer to case reports supporting this theory, as controlled studies investigating percutaneous absorption in relation to the condition of the skin were not available The differences we found for the total group in the relationship of internal to external exposure between persons with healthy skin and those with suspected skin damage support the hypothesis that every pathological skin condition promotes dermal absorption Considering individual subgroups (those with skin diseases, persons with skin irritation, workers in the spinning rooms) the differences were no longer significant but did reveal a similar trend, again supporting the hypothesis. On the basis of our findings the following conclusions may be drawn: 1 Deviations from the normal weight cause a systematic change in the TTCA excretion This influence is noteworthy in considerably overweight employees with exposure at the level of the currently valid threshold

limit values, as in these cases there is a risk of underestimating the internal exposure. The day of the shift, the total length of exposure, age, nicotine abuse and alcohol consumption have no recognisable influence on the TTCA excretion in the post-shift urine. 2 A heavy physical work load and a high dermal absorption rate influence significantly the internal exposure A valid determination of the individual effective dose of the hazardous substance is therefore only possible by biological monitoring. 3 Persons with skin diseases or skin irritations absorb more C 5 2 via the skin than persons with healthy skin Exclusive determination of the external exposure therefore leads to an underestimation of the hazard to these persons. Acknowledgements We particularly wish to thank Mr Miller for his assistance in the collection of air samples and analysis We further thank Dr Freudlsperger, who gave us much support with his practical knowledge and active help This study was carried out with the financial support of the Koelsch-Stiftung e V.

References Aalto-Korte K, Turpeinen M (1993) Transepidermal water loss and absorption of hydrocortisone in widespread dermatitis Br J Dermatol 128:633-635 Beauchamp RO, Bus JA, Popp JA, Boreiko CJ, Goldberg L (1983) A critical review of the literature on carbon disulfide toxicity. CRC Crit Rev Toxicol 11:169-278 Berner B, Cooper ER (1987) Models of skin permeability In: Kydonieu AF, Berner B (eds) Transdermal delivery of drugs vol II CRC Press, Boca Raton, pp 41-55 BUA-Advisory Committee of the GDCh on Existing Chemicals of Environmental Relevance (1993) Carbon disulfide BUA report 83 S Hirzel, Stuttgart Campell L, Jones AH (1985) Wilson KH (1985) Evaluation of occupational exposure to carbon disulfide by blood, exhaled air and urine analysis Am J Ind Med 8:143-153 Cohen AE, Paulus HJ, Keenan RC, Scheel LD (1958) Skin absorption of carbon disulphide vapor in rabbits Arch Ind Health 17:164-169 Drexler H, Goen T, Angerer J, Abou-el-ela S, Lehnert G (1994) Carbon disulphide Part I External and internal exposure to carbon disulphide with workers in the viscose industry Int Arch Occup Environ Health 65:359-365 Dutkiewicz T, Baranowska B (1967) The significance of absorption of carbon disulfide through the skin in the evaluation of exposure In: Brieger H, Teisinger J (eds), Toxicology of carbon disulfide Excerpta Medica Foundation, Amsterdam, pp 50-51 Fartasch M, Bassukas ID, Diepgen TL (1992) Disturbed extruding mechanism of lamellar bodies in dry non-eczematous skin of atopics Br J Dermatol 127:221 227 Freudlsperger FP, Madaus WP (1989) Erfahrungen mit dem BATWert ffir Schwefelkohlenstoff Arbeitsmed Sozialmed Praventivmed 24:71 74 Grandjean P (1990) Skin penetration-hazardous chemicals at work Taylor & Francis, London, pp 171-173 Lavrijsen APM, Oestmann E, Hermans J, Bodde HE, Vermeer BJ, Ponec M (1993) Barrier function parameters in various keratinization disorders-transepidermal water loss and vascular response to hexyl nicotinate Br J Dermatol 129:547-554

10 Rosier J, Vanhoorne M, Grosjean M, Van De Walle E, Billemont G, Van Peteghem C (1982) Preliminary evaluation of urinary 2thio-thiazolidine-4-carboxylic-acid (TTCA) levels as a test for exposure to carbon disulfide Int Arch Occup Environ Health 51:159-167 Rosier J, Veulemans H, Masschelein R, Vanhoorne M, Van Peteghem C (1987) Experimental human exposure to carbon disulfide II Urinary excretion of 2-thiothiazolidine-4-carboxylic acid (TTCA) during and after exposure Int Arch Occup Environ Health 59:243-250

Sachs L (1992) Angewandte Statistik Springer, Berlin, Heidelberg New York Van Poucke L, Van Peteghem C, Vanhoorne M (1990) Accumulation of carbon disulphide metabolites Int Arch Occup Environ Health 62:479-482 World Health Organization (WHO) (1979) Environmental health criteria 10 Carbon disulfide World Health Organization, Geneva