You are here:
NEUROBEHAVIORAL EVALUATIONS OF BINARY AND TERTIARY MIXTURES OF CHEMICALS: LESSIONS LEARNING.
Citation:
Moser, V C., R C. MacPhail, AND C. Gennings. NEUROBEHAVIORAL EVALUATIONS OF BINARY AND TERTIARY MIXTURES OF CHEMICALS: LESSIONS LEARNING. Presented at Symposium on the Neurotoxicity of Mixtures: Aggregate and Cumulative Exposure by Common and Separate Mechanisms, Washington, DC, April 27, 2000.
Description:
The classical approach to the statistical analysis of binary chemical mixtures is to construct full dose-response curves for one compound in the presence of a range of doses of the second compound (isobolographic analyses). For interaction studies using more than two chemicals, response-surface methodology requiring factorial designs is frequently used. The objective of this analysis is to estimate the complex association between the chemicals and the response using a statistical model appropriately parameterized with "interaction parameters." Both of these methods have shortcomings, especially in the number of animals required. Furthermore, interpretation may be complicated when the nature of the interaction among chemicals may be dose-dependent. A full-factorial study was undertaken to determine the neurobehavioral consequences of exposure to three toxicants, in order to characterize potential two- and three-way interactions. Adult female rats were dosed for 10 days with one of five doses of trichloroethylene, diethylhexyl phthalate, or heptachlor, as well as all possible toxicant combinations. Neurobehavioral evaluations were conducted using motor activity and an abbreviated functional observational battery. Response-surface analysis was applied to each of the endpoints. Hypotheses were tested based on the estimated model parameters, and of primary interest was the overall test for interaction among the three toxicants. In addition, an abbreviated design was created by fitting only a subset of the data to the model. In general, significant overall interactions were detected for the majority of the endpoints. Most of these interactions could be characterized as synergism or potentiation. The results of the abbreviated dataset analysis often did not reflect the same outcomes. This study was extremely intensive, in terms of the numbers of rats and time required for conduct of the study and the data analysis. These results underscore the need for more