prior exposure to silica in previous jobs and misclassification of renal diseases may have limited the interpretability of their results. McGeoghegan and Binks (2001) found a nonsignificant increase in deaths due genitourinary diseases in radiation workers compared with the English and Welsh populations (5 observed vs 4.63 expected; SMR, 108) in a study of processors at the British Nuclear Fuels Chapelcross site.
Cragle and colleagues (1988) reported statistically significantly fewer deaths due to genitourinary diseases in hourly employees (SMR, 39; 95% CI, 10-96) in a study of workers at the Savannah River plant. McGeoghegan and Binks (2000b) also reported significantly fewer deaths than expected in radiation workers (SMR, 57; p < 0.01). Frome and colleagues (1997) reported fewer deaths than expected from diseases of the genitourinary system (SMR, 83) in white men in a study of processing workers at the four Federal nuclear plants in Oak Ridge, Tennessee. An earlier study of Oak Ridge workers at the Y-12 and K-25 uranium-enrichment facilities revealed no difference between the numbers of observed and expected deaths from chronic nephritis (SMR, 99; 95% CI, 71-1261) (Frome et al., 1990), as reported in Volume 1. The observed findings were probably influenced by a healthy-worker effect.
In many of the 14 studies, the computed death rates included all genitourinary conditions instead of focusing on renal diseases. Despite reported increases in observed deaths, the SMRs may not have reflected a true response to uranium exposure. In several of the plants, uranium exposure coexisted with other relevant heavy-metal or chemical exposure. Generally, most researchers were unable to isolate the effects of uranium exposure alone.
McDiarmid and colleagues conducted a medical investigation of Gulf War veterans who inhaled or ingested airborne depleted-uranium particles or experienced depleted-uranium wound contamination as a result of friendly-fire incidents and found renal-function measurements that were generally within normal clinical limits (see Table 8-14) (McDiarmid et al., 2000, 2001, 2004, 2006, 2007). Urinary uranium excretion was used in the exposure assessment, and subjects were separated into high- and low-exposure groups on the basis of a cutpoint of 0.10 μg/g of creatinine. In the first of the Baltimore Veterans Affairs Medical Center (BVAMC) studies, veterans with retained depleted-uranium shrapnel fragments had higher urinary uranium concentrations than those without 7 years after first exposure. Urinary uranium ranged from 0.01 to 30.74 μg/g of