A Case Study of the NCRP 156 Wound Model of Embedded DU Using Data From Urine Uranium Concentrations of Wounded Veterans.

Depleted uranium (DU) munitions were initially used by the United States (U.S.) military during the first Persian Gulf War in 1991 in order to penetrate heavily armored vehicles. However, as a result of friendly fire, several U.S. military personnel received intakes from DU munitions. One of the ongoing concerns for these wounded veterans is the potential long-term exposure received from DU embedded fragments. The United States Army Institute of Public Health (AIPH) is the first laboratory that analyzes the urine bioassays from Army Soldiers that are injured with DU fragments. The United States Air Force School of Aerospace Medicine also evaluates bioassays from DU injuries. The urine bioassay data collected by AIPH was evaluated using the NCRP 156 wound model coefficients for the DU-Wafer, Fragment, and Particle models. The maximum likelihood method was used in the Integrated Modules for Bioassay Analysis (IMBA-PPAE) to calculate the estimates of intake and tissue doses. Evaluating the three models for wound retention, the DU-Wafer and Fragment model yielded a credible fit to the bioassay data. Comparing the two models, the DU-Wafer model fits the data better than the Fragment model when comparing their autocorrelation coefficient and chi-squared values of (P 1.73 × 10, c 4.83 × 10), (P 2.01 × 10, c 1.09), respectively. This evaluation supports the validity of both the DU-wafer model as well as the default fragmentation model proposed by NCRP 156.

Application of NCRP 156 Wound Model and ICRP 67 Systemic Plutonium Model for Analysis of Urine Data from Simulated Wounds in Nonhuman Primates.

The predictions of the wound model described in NCRP Report No. 156, coupled with the systemic model described in ICRP 67, were compared with the actual urinary excretion data and wound retention data from nonhuman primates injected intramuscularly or subcutaneously with Pu(IV) citrate. The results indicated that the early behavior of Pu(IV) citrate in wounds can be adequately described by the default retention parameters for moderately retained radionuclides suggested by the report. The urinary excretion rates after 200 d post intake could not be described well by the parameters of any of the default wound models because of the differences in the systemic handling of plutonium by humans compared to nonhuman primates.

Biokinetics of 90Sr in Male Nonhuman Primates.

The current study tests the hypothesis that the biokinetics of Sr can be represented by simplification of the ICRP publication 78 Sr model. Default and proposed models were evaluated by their ability to predict injected activity and more thoroughly define the activity residing in the skeleton of rhesus monkeys. The data obtained from studies done by Patricia Durbin and her colleagues at the Lawrence Berkley National Laboratory were used to create a profile of the activity residing in the skeleton at the time of sacrifice. Post mortem data along with periodic whole body count data were used to optimize the biokinetic parameters using the Integrated Modules for Bioassay Analysis (IMBA), Weighted Likelihood Monte-Carlo Sampling (WeLMoS) program to better predict the intake and fit of the bioassay data. Analysis of the default ICRP 78 parameters resulted in an overprediction of activity in the skeleton for a male cohort by as much as 180%. Using Monte Carlo sampling methods, three models were developed and optimized for a composite cohort of male monkeys. Of the three developed models, one model proved to have the best predictive capabilities. The optimized model C obtained for the male cohort was then tested on a validation cohort to test predictive capabilities. Using the optimized model C parameters, the ability to predict activity in the skeleton was improved in comparison to ICRP 78. Prediction of the intake from bioassay data was also improved by a factor of 2 in comparison to ICRP 78. The results suggest that the modified transfer rates of model C could be used as default parameters for biokinetic nonhuman primate modeling and potentially extrapolated to humans.