A repeated measures approach to the detection of the acute behavioral effects of toluene at low concentrations

Acute behavioral effects of solvents are frequently the critical parameter considered for the recommendation of exposure limit values. Studies of toluene's acute effects on animal behavior during exposure have not been particularly sensitive; increasing the number of observations at low concentrations and the number of animals studied is a reasonable approach to increasing the ability to detect small acute effects. The effects of toluene (178-3000 ppm) were determined in 12 rats trained to nose poke on a probabilistic schedule of food delivery in a small exposure chamber that was highly ventilated with air or a toluene test atmosphere. Animals served as their own controls and were exposed to toluene twice a week for 2 hr during behavioral evaluation. At least six determinations of the effects of toluene were made at concentrations of 178, 300, and 560 ppm, and at least three determinations at 1000, 1780, and 3000 ppm. Biphasic concentration-effect, time-effect, and concentration-response functions were observed. Rates of nose poking increased in 4 of 12 animals at 178 ppm and in 9 of 12 animals at 300 ppm; all animals were affected at 560 ppm. The concentration at which a 10% rate increase occurred was determined by regression analysis to be 156 ppm with the lower 95% confidence limit or "benchmark" concentration being 116 ppm. Weighted regression analysis produced slightly higher estimates of these parameters, 182 and 157 ppm, respectively, while conventional logistic regression revealed that 10% of animals would be affected at a concentration of 114 ppm with a lower confidence limit of 63 ppm. These studies demonstrate that effects on the conditioned performance of rats occur at exposure levels below 300 ppm, the concentration at which robust concentration-related behavioral effects are reported during human experimental exposures to toluene. The statistical methods here have general application in noncancer safety evaluation.