Biological monitoring of vehicle mechanics and other workers exposed to low concentrations of benzene

 It has been suggested that the threshold limit value (TLV) for the time-weighted average (TWA), of benzene be lowered because of its possible leukemogenic effect at low exposure concentrations. This requires the development of new methods of biological monitoring. In this cross-sectional study the diagnostic power of blood and breath benzene and of urinary phenol, catechol, hydroquinone, S-phenylmercapturic acid, and muconic acid were compared in a population of 410 male workers exposed to benzene in garages, in two coke plants, and in a by-product plant. Benzene exposure was assessed by personal air sampling (charcoal tube and passive dosimeter). In all, 95% of the workers were exposed to less than 0.5 ppm benzene. According to the multiple regression equation, the muconic acid and S-phenylmercapturic acid concentrations detected in nonsmokers exposed to 0.5 ppm benzene were 0.3 mg/g and 6 μg/g, respectively (range 0.2–0.6 mg/g and 1.2–8.5 μg/g, respectively). With muconic acid very few false-positive test results were found, and this determination remained reliable even around a cutoff level of 0.1 ppm benzene. Moreover, the diagnostic power of this test proved to be good even when diluted or concentrated urine samples were not excluded. S-Phenylmercapturic acid (S-PMA) also performed fairly well. Blood and breath benzene as well as urinary phenol (PH) and hydroquinone (HQ) were clearly less suitable biomarkers than muconic acid (MA). Catechol (CA) was not associated with occupational benzene exposure. According to the results of biological monitoring, the skin resorption of benzene from gasoline or other fuels seems negligible. Correlation, multiple regression, and likelihood ratios consistently showed that MA and S-PMA concentrations were fairly good indicators of benzene exposure in the 0.1- to 1-ppm range, even in a population comprising both smokers and nonsmokers. PH, HQ, CA, and blood and breath benzene were less suitable, if at all, in the same exposure range. Received: 31 July 1996/Accepted: 29 November 1996