A cohort of seventy-four 1991 Gulf War soldiers with known exposure to depleted uranium (DU) resulting from their involvement in friendly-fire incidents with DU munitions is being followed by the Baltimore Veterans Affairs Medical Center. Biennial medical surveillance visits designed to identify uranium-related changes in health have been conducted since 1993. On-going systemic exposure to DU in veterans with embedded metal fragments is indicated by elevated urine uranium (U) excretion at concentrations up to 1,000-fold higher than that seen in the normal population. Health outcome results from the subcohort of this group of veterans attending the 2005 surveillance visit were examined based on two measures of U exposure. As in previous years, current U exposure is measured by determining urine U concentration at the time of their surveillance visit. A cumulative measure of U exposure was also calculated based on each veteran's past urine U concentrations since first exposure in 1991. Using either exposure metric, results continued to show no evidence of clinically significant DU-related health effects. Urine concentrations of retinol binding protein (RBP), a biomarker of renal proximal tubule function, were not significantly different between the low vs. high U groups based on either the current or cumulative exposure metric. Continued evidence of a weak genotoxic effect from the on-going DU exposure as measured at the HPRT (hypoxanthine-guanine phosphoribosyl transferase) locus and suggested by the fluorescent in-situ hybridization (FISH) results in peripheral blood recommends the need for continued surveillance of this population.