Copyright @ 2012 by the Health Physics Society. Unauthorized reproduction of this article is prohibited. Paper MAXIMUM LIKELIHOODANALYSIS OF BIOASSAY DATA FROM LONG-TERM FOLLOW-UP OF TWO REFRACTORY PUO 2 INHALATION CASES Maia Avtandilashvili,* Richard Brey,* and Anthony C. James†‡ AbstractVThe U.S. Transuranium and Uranium Registries’ tissue donors 0202 and 0407 are the two most highly exposed of the 18 registrants who were involved in the 1965 plutonium fire accident at a defense nuclear facility. Material released during the fire was well characterized as ‘‘high fired’’ refrac- tory plutonium dioxide with 0.32-Km mass median diameter. The extensive bioassay data from long-term follow-up of these two cases were used to evaluate the applicability of the Human Respiratory Tract Model presented by International Commis- sion on Radiological Protection in Publication 66 and its revi- sion proposed by Gregoratto et al. in order to account for the observed long-term retention of insoluble material in the lungs. The maximum likelihood method was used to calculate the point estimates of intake and tissue doses and to examine the effect of different lung clearance, blood absorption and systemic models on the goodness-of-fit and estimated dose values. With appropriate adjustments, Gregoratto et al. particle transport model coupled with the customized blood absorption para- meters yielded a credible fit to the bioassay data for both cases and predicted the Case 0202 liver and skeletal activities mea- sured postmortem. PuO 2 particles produced by the plutonium fire are extremely insoluble. About 1% of this material is absorbed from the respiratory tract relatively rapidly, at a rate of about 1 to 2 d j1 (half-time about 8 to 16 h). The re- mainder (99%) is absorbed extremely slowly, at a rate of about 5  10 j6 d j1 (half-time about 400 y). When considering this situation, it appears that doses to other body organs are negli- gible in comparison to those to tissues of the respiratory tract. About 96% of the total committed weighted dose equivalent is contributed by the lungs. Doses absorbed by these workers’ lungs were high: 3.2 Gy to AI and 6.5 Gy to LN TH for Case 0202 (18 y post-intake); 3.2 Gy to AI and 55.5 Gy to LN TH for Case 0407 (43 y post-intake). This evaluation supports the Gregoratto et al. proposed revision to the ICRP 66 model when considering situations of extremely insoluble particles. Health Phys. 103(1):00Y00; 2012 Key words: analysis, statistical; bioassay; intake, radionuclide; lungs, human INTRODUCTION THE UNITED STATES Transuranium and Uranium Registries (USTUR) represent a unique resource of the human data from U.S. Department of Energy (DOE) employees who were exposed to measurable internal actinide contami- nation in the course of working within the DOE complex. These data are currently used extensively for quantifying the variability in behavior of uranium and transuranic elements among individuals and validating the biokinetic models. With regard to internal exposure, plutonium repre- sents one of the more important actinide elements. Since its discovery, this radionuclide has been extensively used in military and civilian activities associated with weapons development and electrical energy generation. During nuclear fuel production or reprocessing, inhalation is the most common potential pathway of intake. Hence, the quantification of internal doses due to plutonium exposure via the inhalation pathway is of great concern. The Human Respiratory Tract Model (HRTM) ( F1 Fig. 1) presented by the International Commission on Radiologi- cal Protection (ICRP) in Publication 66 (1994a) has dem- onstrated merit under a broad set of situations. However, the ICRP is currently reviewing the applicable data and experimental evidence obtained since publication of the HRTM in order to update the existing model and reduce the uncertainties in internal dosimetry evaluations. To account for the observed long-term retention of insoluble material in the lungs, Gregoratto et al. (2010) proposed a physiologically-based particle transport model that significantly simplifies the representation of parti- cle clearance from the alveolar-interstitial (AI) region by www.health-physics.com 1 *Department of Nuclear Engineering and Health Physics, Idaho State University, 921 South 8th Avenue, Stop 8060, Pocatello ID 83209- 8060; †U.S. Transuranium and Uranium Registries (retired), College of Pharmacy, Washington State University, 1845 Terminal Drive, Suite 201, Richland, WA 99354-4959; ‡Posthumous. AQ1 The authors declare no conflict of interest. For correspondence contact: Maia Avtandilashvili, Department of Nuclear Engineering and Health Physics, Idaho State University, 921 South 8th Avenue, Stop 8060, Pocatello ID 83209-8060, or e-mail at avtamaia@isu.edu. (Manuscript accepted 11 January 2012) 0017-9078/12/0 Copyright * 2012 Health Physics Society DOI: 10.1097/HP.0b013e31824ac627