ORIGINAL ARTICLE Elaine Symanski á Enrico Bergamaschi á Antonio Mutti Inter- and intra-individual sources of variation in levels of urinary styrene metabolites Received: 10 July 2000 /Accepted: 28 December 2000 Abstract Objective: Given the paucity of studies that have examined variability in biological measures of ex- posure to workplace contaminants, we quanti®ed the intra- and inter-individual sources of variation in uri- nary levels of mandelic acid MA) and phenylglyoxylic acid PGA) among workers exposed to styrene. A sec- ondary objective was to examine eects of job task and the timing of sampling during the workweek on the variation in workers' urinary styrene metabolite levels. Methods: As part of routine biological monitoring, a total of 1,714 measurements of MA and PGA collected from 331 workers between 1985 and 1999 from eight reinforced-plastics plants were abstracted from labora- tory reports. To evaluate sources of variation in levels of urinary styrene metabolites, we applied random-eects models. The in¯uence of job task and day of sampling on metabolite levels was examined using mixed-eects models. Results: PGA levels were characterized by less variation than levels of MA, as were metabolite levels expressed in terms of urinary creatinine concentration. The relative magnitude of the inter-individual to the intra-individual source of variation was generally higher for post-shift urine samples than for pre-shift urine samples. As expected, urinary metabolite levels were highest for laminators and for samples collected at the latter end of the workweek. Owing to the eects of variation from day-to-day, estimates of workers' expo- sures that rely on single measurements would generally perform poorly in a regression analysis designed to ex- amine eects resulting from chronic exposure. However, the bias in an observed slope coecient would be diminished if a second or third urine sample were collected. Conclusions: Quanti®cation of the intra- and inter-individual sources of variation provides useful information that can be used to design optimal sampling strategies, which would allow for the collection of sucient data to estimate workers' exposures reliably when evaluating health risks associated with occupa- tional contaminants. Key words Exposure assessment á Styrene á Mandelic acid á Phenylglyoxylic acid á Biological monitoring Introduction Styrene is an important chemical used in a broad range of industrial applications including the manufacture of plastics, synthetic rubbers, and resins. Airborne levels of styrene are generally low in most industrial settings ex- cept for the reinforced-plastics industry World Health Organization 1983; IARC 1994), which uses styrene as a solvent for unsaturated resin and as a reactant for po- lymerization during the production of boats, tanks, and other reinforced-plastics products. The main route of uptake of styrene is through inhalation; studies suggest that percutaneous absorption contributes negligible amounts to total exposure in the reinforced-plastics industry Sorsa et al. 1991; Limasset et al. 1999). Ap- proximately 60 to 70% of the inhaled styrene is retained in the body and is readily distributed to the blood and fat tissues Bond 1989; Sumner and Fennell 1994). The major metabolic pathway involves the oxidation of sty- rene to styrene-7,8-oxide by hepatic cytochrome P-450 isozymes. Styrene oxide is hydrolyzed to styrene glycol, which is subsequently oxidized into mandelic acid MA) and further to phenylglyoxylic acid PGA). Excretion of MA appears to be biphasic, with half-times that have been reported to range from 4 to 9 h for the fast elimi- nation phase and 17 to 25 h for the slow elimination phase Bond 1989). Although there is some indication Int Arch Occup Environ Health 2001) 74: 336±344 Ó Springer-Verlag 2001 E. Symanski &) University of Texas School of Public Health, 1200 Herman Pressler Street, Houston, TX 77030, USA e-mail: esymanski @sph.uth.tmc.edu Tel.: +1-713-500-9238; Fax: +1-713-500-9249 E. Bergamaschi á A. Mutti Laboratory of Industrial Toxicology, Department of Clinical Medicine, Nephrology and Health Sciences, University of Parma Medical School, Parma, Italy