Comparison of Hippuric Acid and O-Cresol in Urine and Unchanged Toluene in Alveolar Air for the Biological Monitoring of Exposure to Toluene in Human Volunteers

Abstract

Toluene (TOL) is one of the most widely used organic solvents in the workplace. Recently, the permissible exposure limit to TOL has been reduced from 100 to 50 ppm by the American Conference of Governmental Industrial Hygienists. Urinary hippuric acid (HA) is not recommended for monitoring the exposure to low ambient concentrations of TOL. Despite the fact that there are other potential indicators of exposure to TOL, there is still no consensus on which one is the most reliable for use on a large scale. This study was undertaken to describe how three indicators of TOL exposure—namely, HA and o-cresol (o-CR) in urine—and unchanged TOL in alveolar air (TOL-A) relate to graded exposure levels of TOL in human volunteers. In addition, the potential impact of simultaneous exposure to xylene (XYL) and TOL exposure on indicators of exposure to TOL was investigated in view of the known metabolic interference between these two solvents. Four adult nonsmoking volunteers were exposed (7 hours) to various concentrations of TOL (10, 20, 30, 50, and 100 ppm). On three other occasions, the volunteers were exposed to three different binary mixtures containing TOL and XYL. The concentrations of HA and o-CR were determined in urine samples collected during and after exposures, whereas TOL was measured in end-exhaled (alveolar) air samples collected at 30-minute intervals during and after exposure. There was a good correlation between TOL exposure and the urinary levels of HA, especially in samples collected during the last 4 hours of exposure (3 to 7 hours) (r = 0.87). The urinary levels of o-CR were also well correlated with TOL exposure for samples collected during the last 4 hours of exposure (r = 0.81) and for samples collected during the 17 hours following exposure (r = 0.88). However, at low exposure concentrations of TOL (i.e., 10 ppm), the excretion (3 to 7 hours) of HA (e.g., 0.23 ± 0.02 mmol/mmol creatinine) was only slightly above the mean background level measured before exposure (0.22 ± 0.06 mmol/mmol creatinine), while that of o-CR was clearly above background (0.12 ± 0.03 versus 0.04 μmol/mmol creatinine in one subject). TOL-A was strongly correlated with exposure concentrations of TOL when measured at the end (r = 0.99), or even 30 minutes after cessation of exposure (r = 0.97). Results of the present study confirm that HA cannot be used to monitor low exposure to TOL and suggest that o-CR and TOL-A are sensitive enough to be used for that purpose. Among three indicators of exposure, only o-CR was significantly affected (reduction of ∼20 to 30% in amount excreted) by simultaneaous exposure to XYL at a total mixture concentration not exceeding approximately 100 ppm, whereas HA and TOL-A remained unaltered. In conclusion, this study shows that TOL-A and o-CR can be used to monitor exposure to low levels of TOL. It also shows that, between these two indicators, o-CR may be especially sensitive to metabolic interferences resulting from coexposure to other solvents sharing similar metabolic pathways, even at exposure levels below exposure limits for mixtures of solvents.