Apparent half-lives of dioxins, furans, and polychlorinated biphenyls as a function of age, body fat, smoking status, and breast-feeding

Abstract

Objective: In this study we reviewed the half-life data in the literature for the 29 dioxin, furan, and polychlorinated biphenyl congeners named in the World Health Organization toxic equivalency factor scheme, with the aim of providing a reference value for the half-life of each congener in the human body and a method of half-life estimation that accounts for an individual's personal characteristics.

Data sources and extraction: We compared data from >30 studies containing congener-specific elimination rates. Half-life data were extracted and compiled into a summary table. We then created a subset of these data based on defined exclusionary criteria.

Data synthesis: We defined values for each congener that approximate the half-life in an infant and in an adult. A linear interpolation of these values was used to examine the relationship between half-life and age, percent body fat, and absolute body fat. We developed predictive equations based on these relationships and adjustments for individual characteristics.

Conclusions: The half-life of dioxins in the body can be predicted using a linear relationship with age adjusted for body fat, smoking, and breast-feeding. Data suggest an alternative method based on a linear relationship between half-life and total body fat, but this approach requires further testing and validation with individual measurements.

Keywords: elimination rate; half-life; pharmacokinetics; poly-chlorinated dibenzofurans; polychlorinated biphenyls; polychlorinated dibenzo-p-dioxins.

Figure 1

Range of half-life values (in years) for dioxins and furans based on a subset of values from the literature. Bars represent 25th, 50th, and 75th percentiles, and whiskers indicate the range. Diamonds indicate the reference values within this range, circles indicate outliers, and asterisks indicate extreme cases.

Figure 2

Range of half-life values (in years) for PCBs based on a subset of values from the literature. Bars represent 25th, 50th, and 75th percentiles, and whiskers indicate the range. Diamonds indicate the reference values within this range, circles indicate outliers, and asterisks indicate extreme cases.

Figure 3

Half-life of TCDD as a function of age. aApplication of the model presented by Van der Molen et al. (2000) to the Flesch-Janys et al. (1996) data as done by Ogura (2004). bValues from the current literature presented in Table 1. cLinear interpolation between the infant and adult reference half-lives (slope and intercept given in Table 5).

Figure 4

Half-life of 1,2,3,7,8-PeCDD as a function of age. Application of the equation proposed by Flesch-Janys et al. (1996) for ages > 60 years may be problematic because very small variations in the elimination rate could lead to substantial divergence in half-life length. aApplication of the model presented by Van der Molen et al. (2000) to the Flesch-Janys et al. (1996) data as done by Ogura (2004). bValues from the current literature presented in Table 1. cLinear interpolation between the infant and adult reference half-lives (slope and intercept given in Table 5).

Figure 5

TCDD half-life as a function of percent body fat. The oval indicates the area where the relationship of increased half-life with increased body fat does not hold; these values represent young subjects. Literature-reported data enclosed in squares indicate subjects whose half-lives were measured when they had serum concentrations that were well above the level of increased induction of degradation enzymes. aApplication of the model presented by Van der Molen et al. (2000) to the Flesch-Janys et al. (1996) data as done by Ogura (2004). bValues from the current literature presented in Table 1.

Figure 6

TCDD half-life as a function of total body fat. The two points shown in the square represent subjects whose half-lives were measured when they had serum concentrations well above the level of increased induction of degradation enzymes. aApplication of the model presented by Van der Molen et al. (2000) to the Flesch-Janys et al. (1996) data as done by Ogura (2004). bValues from the current literature presented in Table 1.