A study employing several biomarkers of styrene exposure and genotoxicity was carried out in a group of lamination (reinforced plastic) workers and controls, who had been repeatedly sampled during a 3-year period. Special attention will be paid to the last sampling (S.VI), reported here for the first time. Styrene concentration in the breathing zone, monitored by personal dosimeters, and urinary mandelic acid (MA) were measured as indicators of external exposure. Blood samples were assayed for styrene-specific O6-guanine adducts in DNA, N-terminal valine adducts of styrene in haemoglobin, DNA single-strand breaks (SSB), determined by use of the single cell gel electrophoresis (Comet) assay), and hypoxanthine guanine phosphoribosyl transferase (HPRT) mutant frequencies (MF) in T-lymphocytes. O6-styrene guanine adduct levels were significantly higher in the exposed group (5.9 +/- 4.9 adducts/10(8) dNp) as compared to laboratory controls (0.7 +/- 0.8 adducts/10(8) dNp; P = 0.001). DNA adduct levels significantly correlated with haemoglobin adducts, SSB parameters and years of employment. Styrene-induced N-terminal valine adducts were detected in the lamination workers (1.7 +/- 1.1 pmol/g globin), but not in the control group (detection limit 0.1 pmol/g globin). N-terminal valine adducts correlated strongly with external exposure indicators, DNA adducts and HPRT MF. No significant correlation was found with SSB parameters. A statistically significant difference in HPRT MF was observed between the laminators (22.3 +/- 10.6/10(6)) and laboratory controls (14.2 +/- 6.5/10(6), P = 0.039). HPRT MF in the laminators significantly correlated with styrene concentration in air, MA and haemoglobin adducts, as well as with years of employment and age of the employees. No significant difference (P = 0.450) in MF between the laminators and the factory controls was observed. Surprisingly, we detected differences in MF between sexes. When data from all measurements were combined, women showed higher MF (geometric mean 15.4 vs. 11.2 in men, P = 0.020). The styrene-exposed group exhibited significantly higher SSB parameters (tail moment (TM), tail length (TL) and the percentage of DNA in the tail (TP)) than the control group (P < 0.001). SSB parameters correlated with indicators of external exposure and with O6-styrene guanine adducts. No significant correlation was found between SSB parameters and haemoglobin adducts or HPRT MF. The data encompassing biomarkers from repeated measurements of the same population over a 3-year period are discussed with respect to the mechanisms of genotoxic effects of styrene and the interrelationship of individual biomarkers.