Abstract:
WIth the increase in application of chemicals in industrial processes the trend of industrial hygiene management has grown rapidly. For the control of workplace hazards occurring form chemical exposure that can be potentially hazardous to workers, practice of exposure risk assessment tools has been established. Recently this domain of occupational health has gained increased attention globally in industries with more stringent regulations being introduced for workers safety from chemical exposures. The rigorous scientific computational method employed to simulate such exposure are the physiologically based pharmacokinetic (PBPK) models, making the risk assessment process more efficient and rapid. These simulations are based on the relationship between chemical specific parameters and organisms related biological properties. This project proposes a model for chemical blood: air partition coefficient that can be incorporated in PBPK models. The experimental values of blood: air partition coefficient required for assessments are available for a limited number of chemicals and are resource intensive. Similarly, the existing estimation models involve many parameters which are not readily available and are comparatively a much more expensive approach than our proposed model. With a linear combination of only two parameters as input, partition coefficient of air-water and octanol-water, our model provides results that are validated and consistent with existing complex model. Inhalation is the major pathway of exposure for most chemicals, but this study focuses mainly on volatile organic compounds present in occupational settings that easily escape into air from several industrial processes and get systemically absorbed into blood. Therefore, exposure assessment will be more accurate if blood VOCs concentrations are analyzed instead of air concentrations and so this model brings a simple and computationally inexpensive way to simulate occupational exposure to VOCs. The proposed model successfully explains variation in the blood-air distribution data (n =196) with 0.93 and root mean square error (RMSE) = 0.324 log units. The evaluation of developed model was assessed in comparison to existing Abraham
Solvation Model (ASM) through case study for industrial chemicals (n = 65) resulting in RMSE = 0.29 log units for our model and RMSE = 0.29 log units for ASM, indicating that the proposed model performs well and can be used as a substitute for complex model.
