METHODS

Collectives

A total of 62 employees (7 females, 55 males) of the chemical plant producing surfactants for textile industry took part in this study; 38 were smokers, and 24 were non-smokers. The age of the workers was 19–59 years with a median age of 35 years. All of the employees stated that they wear protective gloves during their work.

Eight workers in a hot spot area (as revealed by the first investigation) took part in a follow up investigation four months later. The median age of this subgroup (four smokers, four non-smokers) was 36 years (range 28–54 years).

The alkylating substances (ethylene oxide, propylene oxide, or dimethyl sulphate) processed in the plant are pumped via pipeline into sealed vessels for reaction with various compounds (for example, fatty acids, phenols, or alcohols). The reaction products are then finally filled into barrels or containers.

Stationary air monitoring in this area has been conducted regularly by the company itself as well as by the German Berufsgenossenschaft using an official GC/MS method with a limit of detection of 10 µg dimethyl sulphate/m³.⁸

The control group consisted of 10 persons (2 females, 8 males) from our laboratory with no exposure to dimethyl sulphate or other methylating agents. Two of these persons were smokers, eight were non-smokers. The median age of the control group was 30 years(range 27–59 years).

Adduct monitoring

The biological adduct monitoring was conducted according to the method recommended by the Deutsche Forschungsgemeinschaft.⁹ After isolation of erythrocytes and globin from the blood, the level of N-methylvaline was quantified using the modified Edman procedure with N-methylvaline-leucine-anilide as calibration standard and N-ethoxy­ethyl­valine-alanine-anilide as internal standard. The formed pentafluorophenyl thiohydantoine derivatives were finally determined and quantified with mass spectrometric detection (EI mode) after gas chromatographic separation on a DB-17 HT capillary column (crosslinked (50%-phenyl)-methylpolysiloxane, 30 m×0.25 mm ID, 0.15 µm film thickness, J & W Scientific, Folsom, CA, USA).

The resulting data were expressed as micrograms adduct per litre blood, as the German exposure equivalent is also expressed in this unit. For comparison with other determinations, 10 µg N-methylvaline/l blood is considered equivalent to 525 pmol adduct/g globin.

For quality control purposes, blood of a smoker and a non-smoker has been included in each analytical series. The between series precision (n = 8) has been determined to be 7.9% for the smoker (mean concentration 12.2 µg/l blood) and 6.9% for the non-smoker (mean concentration 9.2 µg/l blood).

RESULTS

Table 1 summarises the results. The internal exposure of a group of workers is illustrated in the relative cumulative frequency for N-methylvaline depicted in fig 1. Figure 2 presents the results of our follow up study of eight workers with high internal exposure to dimethyl sulphate.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Relative cumulative frequency for N-methylvaline in blood of workers and controls.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

N-methylvaline levels of eight workers with high internal burden in the first investigation compared to the follow up four months later. The broken line indicates the EKA value for dimethyl sulphate of 40 µg/l blood.

DISCUSSION

As shown in fig 1, 52 of the 62 workers (85%) showed N-methylvaline levels in the range of the general population that do not differ significantly (p = 0.1, Wilcoxon test, two tailed) from the values obtained from our control group (especially regarding the different smoking habits). However, our study revealed a massive exposure to dimethyl sulphate in a group of 10 employees working in a specific building of the chemical plant, where this substance is processed. This shows that the exposure is strictly localised in this area. The values for N-methylvaline in blood of these workers even exceeded the exposure equivalent value for dimethyl sulphate up to fourfold (see fig 1).

These results were quite surprising, because stationary air monitoring of dimethyl sulphate in this area performed both by the company itself as well as by official authorities showed only values below the limit of detection (10 µg/m³). This obviously led to an underestimation of the actual burden of the workers.

As a possible reason for these high exposures, skin contact with residues of unreacted dimethyl sulphate in the product was assumed. Since this hot spot could be identified by use of biological monitoring, it was decided to further investigate the internal exposure of these workers in order to control the efficiency of preventive measures.

In the second investigation four months after the first, only 8 of the 10 workers with the highest internal burdens that had already participated in the first study took part. Two employees did not participate in this follow up study for various reasons.

As shown in fig 2, the N-methylvaline levels of this subgroup of workers were still increased in the follow up, but significantly lower for most of the workers (p = 0.05, Wilcoxon test, two tailed). After the results of our first study, the company had taken measures to reduce the exposure. The reaction and stirring times for the product were prolonged in order to diminish dimethyl sulphate residues. Furthermore, a bushing was introduced for the suction tube of the pump so as to minimise skin contact when filling barrels with reaction products. Beyond that, the exhausting possibilities were improved and workers were advised to wear protective helmets with external air supply during opening of the reaction vessel. Apparently, these measures appeared successful in improving the exposure situation of the workers. Thus, the median value for the haemoglobin adduct of dimethyl sulphate in these eight workers decreased from 79.8 µg/l blood to 36.4 µg/l blood in the follow up, although three workers still exceeded the level of the exposure equivalent.

Normal levels for N-methylvaline in the general population have already been reported in the literature.^10–12 The physiological sources of this background level are probably endogenous methylation with S-adenosylmethionine. Smoking habits also had an influence on N-methylvaline levels, apparently due to methylating agents present in tobacco smoke (for example, nitrosamines).¹¹ Most recent publications showed median levels for non-smokers of about 8.5 µg/l blood and for smokers of about 11 µg/l blood, which is in good agreement with the values found in our control group and most of the workers (see table 1).¹²

As far as we are aware, this is the first study to report such excess levels of N-methylvaline after occupational exposure to dimethyl sulphate in a subgroup of workers. Ambient monitoring was insufficient for an estimation of the real exposure levels of workers. The vapour pressure of dimethyl sulphate is quite low; air monitoring of this substance requires highly sensitive methods and has several shortcomings. Dimethyl sulphate may be hydrolysed by air humidity on the sorbent material or may easily react with other compounds present in ambient air (for example, alcohols, amines, etc) and therefore evade quantification.²

The German commission for the investigation of health hazards of chemical compounds in the work area has investigated the relation between the concentration of carcinogens in the workplace air and that of the substance or corresponding adducts in biological material. From these relations, the body burden which results from uptake of the substance exclusively by inhalation may be determined. Basing on this linear relation established for dimethyl sulphate, the maximum value for N-methylvaline found in the blood of the workers would correspond to an inhalative exposure to more than 1000 µg dimethyl sulphate/m³ air.^4,6

Dermal absorption seems to be the most probable route of uptake in this case. We therefore question whether ambient monitoring is still a reasonable way of estimating occupational exposure to dimethyl sulphate. As air monitoring does not reflect all routes of uptake, it may consequently underestimate total exposure and hence suggest a “safe” workplace when this is not the case.

Our results stress the need for biological monitoring of workers handling compounds which are taken up via the skin. Considering the high production volume, the very scarce literature on this compound and the toxicological profile, dimethyl sulphate may indeed represent a “hidden” problem in occupational medicine that deserves closer attention.