Risk estimates for deterministic health effects of inhaled weapons grade plutonium

Risk estimates for deterministic effects of inhaled weapons-grade plutonium (WG Pu) are needed to evaluate potential serious harm to (1) U.S. Department of Energy nuclear workers from accidental or other work-place releases of WG Pu; and (2) the public from terrorist actions resulting in the release of WG Pu to the environment. Deterministic health effects (the most serious radiobiological consequences to humans) can arise when large amounts of WG Pu are taken into the body. Inhalation is considered the most likely route of intake during work-place accidents or during a nuclear terrorism incident releasing WG Pu to the environment. Our current knowledge about radiation-related harm is insufficient for generating precise estimates of risk for a given WG Pu exposure scenario. This relates largely to uncertainties associated with currently available risk and dosimetry models. Thus, rather than generating point estimates of risk, distributions that account for variability/uncertainty are needed to properly characterize potential harm to humans from a given WG Pu exposure scenario. In this manuscript, we generate and summarize risk distributions for deterministic radiation effects in the lungs of nuclear workers from inhaled WG Pu particles (standard isotopic mix). These distributions were developed using NUREG/CR-4214 risk models and time-dependent, dose conversion factor data based on Publication 30 of the International Commission on Radiological Protection. Dose conversion factors based on ICRP Publication 30 are more relevant to deterministic effects than are the dose conversion factors based on ICRP Publication 66, which relate to targets for stochastic effects. Risk distributions that account for NUREG/CR-4214 parameter and model uncertainties were generated using the Monte Carlo method. Risks were evaluated for both lethality (from radiation pneumonitis) and morbidity (due to radiation-induced respiratory dysfunction) and were found to depend strongly on absorbed-dose-rate history (which depends on WG Pu solubility). The results obtained were compared to new data from animal and epidemiological studies. Our findings suggest that NUREG/CR-4214 models for assessing lethality risk from inhaled radionuclides may need some modest revision in light the new data. However, for assessing morbidity risks, major revisions may be needed.