Ischaemic heart and cerebrovascular disease mortality in uranium enrichment workers

Abstract

Objective: Linear and non-linear dose-response relationships between radiation absorbed dose to the lung from internally deposited uranium and external sources and circulatory system disease (CSD) mortality were examined in a cohort of 23 731 male and 5552 female US uranium enrichment workers.

Methods: Rate ratios (RRs) for categories of lung dose and linear excess relative rates (ERRs) per unit lung dose were estimated to evaluate the associations between lung absorbed dose and death from ischaemic heart disease (IHD) and cerebrovascular disease.

Results: There was a suggestion of modestly increased IHD risk in workers with internal uranium lung dose above 1 milligray (mGy) (RR=1.4, 95% CI 0.76 to 2.3) and a statistically significantly increased IHD risk with external dose exceeding 150 mGy (RR=1.3, 95% CI 1.1 to 1.6) compared with the lowest exposed groups. ERRs per milligray were positive for IHD and uranium internal dose and for both outcomes per gray external dose, although the CIs generally included the null.

Conclusions: Non-linear dose-response models using restricted cubic splines revealed sublinear responses at lower internal doses, suggesting that linear models that are common in radioepidemiological cancer studies may poorly describe the association between uranium internal dose and CSD mortality.

Keywords: cardiovascular; mortality studies; radiation.

Fig. 1.

Relative rates (RR) versus absorbed lung dose from internal uranium exposure (A and B) and external ionizing radiation exposure (C and D). Linear model: dashed lines with dose limited to <10 mGy and < 1 mGy for internal dose and dose limited to <500 mGy and <100 mGy for external dose; Categorical model: vertical lines at midpoint of dose categories; Spline model: solid line with equally-spaced knots; 95% confidence interval for spline model: gray shaded area; 99^th percentile of dose: light gray vertical line. External ionizing radiation exposure includes exposure from work-related photofluorographic X-ray examinations administered at the K-25 facility. RRs were evaluated with no lag and adjusted for matching variables: age, sex, race, birth date, facility, and socio-economic status (SES).