The Mayak Worker Dosimetry System (MWDS-2013): Internal Dosimetry Results

Incidence risks for subtypes of heart diseases in a Russian cohort of Mayak Production Association nuclear workers

Heart diseases are one of the main causes of death. The incidence risks were assessed for various types of heart diseases (HDs) in a cohort of Russian nuclear workers of the Mayak Production Association (PA) who had been chronically occupationally exposed to external gamma and/ or internal alpha radiation. The study cohort included all workers (22,377 individuals) who had been hired at the Mayak PA during 1948-1982 and followed up until 31 December 2018. The mean gamma-absorbed dose to the liver (standard deviation) was 0.43 (0.63) Gy, and the mean alpha-absorbed dose to the liver was 0.25 (1.19) Gy. Excess relative risk (ERR) per unit liver-absorbed dose (Gy) was calculated based on maximum likelihood. At the end of the follow-up, 559 chronic rheumatic heart disease (CRHD), 7722 ischemic heart disease (IHD) [including 2185 acute myocardial infarction (AMI) and 3976 angina pectoris (AP)], 4939 heart failure (HF), and 3689 cardiac arrhythmia and conduction disorder (CACD) cases were verified in the study cohort. Linear model fits of the gamma dose response for HDs were best once adjustments for non-radiation factors (sex, attained age, calendar period, smoking status and alcohol consumption) and alpha dose were included. ERR/Gy in males and females was 0.17 (95% confidence intervals: 0.10, 0.26) and 0.23 (0.09, 0.38) for IHD; 0.18 (0.09, 0.29) and 0.26 (0.08, 0.49) for AP; - 0.01 (n/a, 0.1) and - 0.01 (n/a, 0.27) for AMI; 0.27 (0.16, 0.40) and 0.27 (0.10, 0.49) for HF; 0.32 (0.19, 0.46) and 0.05 (- 0.09, 0.22) for CACD; 0.73 (- 0.02, 2.40) and - 0.12 (- 0.50, 0.69) for CRHD, respectively. Sensitivity analyses demonstrated the persistence of a significant dose-response regardless of exclusion/inclusion of adjustments for known potential non-radiation confounders (smoking, alcohol consumption, body mass index, hypertension, diabetes mellitus), and it was only the magnitude of the risk estimate that varied. The risks of HD incidence were not modified with sex (except for the CACD risk). This study provides evidence for a significant association of certain types of HDs with cumulative dose of occupational chronic external exposure to gamma radiation.

Mortality from various diseases of the circulatory system in the Russian Mayak nuclear worker cohort: 1948-2018

This paper reports on the findings from the study of mortality from diseases of the circulatory system (DCS) in Russian nuclear workers of the Mayak Production Association (22 377 individuals, 25.4% female) who were hired at the facility between 1948 and 1982 and followed up until the end of 2018. Using the AMFIT module of the EPICURE software, relative risks (RRs) and excess RRs per unit absorbed dose (ERR/Gy) for the entire Mayak cohort, the subcohort of workers who were residents of the dormitory town of Ozyorsk and the subcohort of migrants from Ozyorsk were calculated based on maximum likelihood. The mean cumulative liver absorbed gamma-ray dose from external exposure was 0.45 (0.65) Gy (mean (standard deviation)) for men and 0.37 (0.56) Gy for women. The mean cumulative liver absorbed alpha dose from internal exposure to incorporated plutonium was 0.18 (0.65) Gy for men and 0.40 (1.92) Gy for women. By the end of the follow-up, 6019 deaths with DCS as the main cause of death were registered among Mayak Production Association workers (including 3828 deaths in the subcohort of residents and 2191 deaths in the subcohort of migrants) over 890 132 (622 199/267 933) person-years of follow-up. The linear model that took into account non-radiation factors (sex, attained age, calendar period, smoking status and alcohol drinking status) and alpha radiation dose (via adjusting) did not demonstrate significant associations of mortality from DCS, ischaemic heart disease (IHD) and cerebrovascular disease with gamma-ray exposure dose in the entire cohort, the resident subcohort or the migrant subcohort (either in men or women). For the subcohort of residents, a significant association with gamma dose was observed for mortality from ischaemic stroke in men with ERR/Gy = 0.43 (95% CI 0.08; 0.99); there were no significant associations with liver absorbed gamma dose for any other considered outcomes. As for internal exposure, for men no significant associations of mortality from any DCS with liver absorbed alpha dose were observed, but for women positive associations were found for mortality from DCS (the entire cohort and the resident subcohort) and IHD (the entire cohort). No significant associations of mortality from various types of DCS with neutron dose were observed either in men or women, although neutron absorbed doses were recorded in only 18% of the workers.

The Mayak Worker Dosimetry System (MWDS-2013): Implementation of the Dose Calculations

The calculation of internal doses for the Mayak Worker Dosimetry System (MWDS-2013) involved extensive computational resources due to the complexity and sheer number of calculations required. The required output consisted of a set of 1000 hyper-realizations: each hyper-realization consists of a set (1 for each worker) of probability distributions of organ doses. This report describes the hardware components and computational approaches required to make the calculation tractable. Together with the software, this system is referred to here as the 'PANDORA system'. It is based on a commercial SQL server database in a series of six work stations. A complete run of the entire Mayak worker cohort entailed a huge amount of calculations in PANDORA and due to the relatively slow speed of writing the data into the SQL server, each run took about 47 days. Quality control was monitored by comparing doses calculated in PANDORA with those in a specially modified version of the commercial software 'IMBA Professional Plus'. Suggestions are also made for increasing calculation and storage efficiency for future dosimetry calculations using PANDORA.

The Mayak Worker Dosimetry System (MWDS-2013): Internal Dosimetry Results

The distribution of calculated internal doses has been determined for 8043 Mayak Production Associate (Mayak PA) workers. This is a subset of the entire cohort of 25 757 workers, for whom monitoring data are available. Statistical characteristics of point estimates of accumulated doses to 17 different tissues and organs and the uncertainty ranges were calculated. Under the MWDS-2013 dosimetry system, the mean accumulated lung dose was 185 ± 594 mGy (geometric mean = 28 mGy; geometric standard deviation = 9.32; median value = 31 mGy; maximum value = 8980 mGy). The ranges of relative standard uncertainty were from 40 to 2200% for accumulated lung dose, from 25-90% to 2600-3000% for accumulated dose to different regions of respiratory tract, from 13-22% to 2300-2500% for systemic organs and tissues. The Mayak PA workers accumulated internal plutonium lung dose is shown to be close to log normal. The accumulated internal plutonium dose to systemic organs was close to a log triangle. The dependency of uncertainty of accumulated absorbed lung and liver doses on the dose estimates itself is also shown. The accumulated absorbed doses to lung, alveolar-interstitial region, liver, bone surface cells and red bone marrow calculated both with MWDS-2013 and MWDS-2008 have been compared. In general, the accumulated lung doses increased by a factor of 1.8 in median value, while the accumulated doses to systemic organs decreased by factor of 1.3-1.4 in median value. For the cases with identical initial data, accumulated lung doses increased by a factor of 2.1 in median value, while accumulated doses to systemic organs decreased by 8-13% in median value. For the cases with both identical initial data and all of plutonium activity in urine measurements above the decision threshold, accumulated lung doses increased by a factor of 2.7 in median value, while accumulated doses to systemic organs increased by 6-12% in median value.

Mayak Worker Dosimetry System (MWDS-2013): Phase I-Quality Assurance of Organ Doses and Excretion Rates From Internal Exposures of Plutonium-239 for the Mayak Worker Cohort

The calculation of reliable and realistic doses for use in epidemiological studies for the quantification of risk from internal exposure to radioactive material is fundamental to the development of advice, guidance and regulations for the control and use of radioactive material. Thus, any programme of work carried out which requires the calculation of doses for use by epidemiologists ideally should contain a rigorous program of quality assurance (QA). This paper describes the initial QA (Phase I) implemented by Public Health England (PHE) and the Southern Urals Biophysics Institute (SUBI) as part of the work programme on internal dosimetry in the Joint Coordinating Committee for Radiation Effects Research Project 2.4 for the 2013 Mayak Worker Dosimetry System. SUBI designed and implemented new software (PANDORA) to include the latest Mayak Worker Dosimetry System and to calculate organ burdens, urinary excretion rates, intakes and absorbed doses, while PHE modified their commercially available IMBA Professional Plus software package. Comparisons of output from the two codes for the Mayak Worker Dosimetry System 2013 showed calculated values of absorbed doses, intakes, organ burdens and urinary excretion agreed to within 1%. The 1% discrepancy can be explained by the approximation used in IMBA to speed up dose calculations.