Toxicology Letters 108 (1999) 303 – 308 A modeling approach to account for toxicokinetic interactions in the calculation of biological hazard index for chemical mixtures S. Haddad a , R. Tardif a , C. Viau a,b , K. Krishnan a,b, * a Groupe de Recherche en Toxicologie Humaine (TOXHUM), Uniersite ´ de Montre ´al, Montre ´al, QC, Canada b E ´ quipe de Recherche en Sureillance Biologique, Uniersite ´ de Montre ´al, Montre ´al, QC, Canada Accepted 19 February 1999 Abstract Biological hazard index (BHI) is defined as biological level tolerable for exposure to mixture, and is calculated by an equation similar to the conventional hazard index. The BHI calculation, at the present time, is advocated for use in situations where toxicokinetic interactions do not occur among mixture constituents. The objective of this study was to develop an approach for calculating interactions-based BHI for chemical mixtures. The approach consisted of simulating the concentration of exposure indicator in the biological matrix of choice (e.g. venous blood) for each component of the mixture to which workers are exposed and then comparing these to the established BEI values, for calculating the BHI. The simulation of biomarker concentrations was performed using a physiologically-based toxicokinetic (PBTK) model which accounted for the mechanism of interactions among all mixture components (e.g. competitive inhibition). The usefulness of the present approach is illustrated by calculating BHI for varying ambient concentrations of a mixture of three chemicals (toluene (5 – 40 ppm), m-xylene (10–50 ppm), and ethylbenzene (10–50 ppm)). The results show that the interactions-based BHI can be greater or smaller than that calculated on the basis of additivity principle, particularly at high exposure concentrations. At lower exposure concentrations (e.g. 20 ppm each of toluene, m-xylene and ethylbenzene), the BHI values obtained using the conventional methodology are similar to the interactions-based methodology, confirming that the consequences of competitive inhibition are negligible at lower concentrations. The advantage of the PBTK model-based methodology developed in this study relates to the fact that, the concentrations of individual chemicals in mixtures that will not result in a significant increase in the BHI (i.e. 1) can be determined by iterative simulation. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Mixtures; Interactions; BEI; BHI; PBPK modeling www.elsevier.com/locate/toxlet * Corresponding author. Postal address: De ´partement de Me ´decine du Travail et d’Hygie `ne du Milieu, Universite ´ de Montre ´al, 2375 Chemin de la Co ˆ te Ste Catherine Bur. 4105, Montre ´al, PQ, H3T 1A8 Canada. Tel.: +1-514-343-6134; fax: +1-514-343-2200. E-mail address: kannan.krishnan@umontreal.ca (K. Krishnan) 0378-4274/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII:S0378-4274(99)00102-2