The skin reservoir effect of [14C]pyrene (in vivo and in vitro) on percutaneous absorption was determined in male Sprague Dawley rats. The urinary 1-OHpyrene (1-OHPy) excretion was compared between dermal exposure and intravenous administration. In vivo, the percutaneous absorption flux of [14C]pyrene (200 microg/cm(2); 50 microL/cm(2) of ethanol) determined by sacrificing batches of rats after different exposure times over 4.5 h was 1.0 +/- 0.1 microg/cm(2) h(-1). During exposure, penetration flux was twofold higher than absorption flux, indicating a gradual accumulation of pyrene in the skin. [14C] skin content at the end of exposure was 16 microg/cm(2), which decreased gradually over time to 2 microg/cm(2) 68 h after the end of exposure. The total absorbed dose during exposure was threefold lower than that after exposure, indicating a high contribution of pyrene skin content to the systemic availability of the compound. Similar results were obtained in vitro. The apparent elimination rate of [14C]pyrene (23 h) contained in the skin after an exposure of 4.5 h was similar to the apparent urinary excretion half life of 1-OHPy (21 h). These values are threefold higher than the urinary excretion half life of 1-OHPy after an intravenous administration of pyrene (0.5 mg/kg). In conclusion, absorbed dose and percutaneous absorption flux were well estimated from the 1-OHPy urinary excretion rate. For risk assessment purposes, the penetration flux rather than the absorption flux should be taken into account for topical pyrene exposure.