A normative value pilot study: levels of uranium in urine samples from UK civilians

A normative study of the levels of urinary uranium in the general UK population is needed for comparison with levels in UK military and ex-military personnel who served where munitions containing depleted uranium (DU) were used. As preparation, this pilot study trialled the process of collecting 24-h samples from adult male civilians, and compared the measurements from 24-h samples with those from spot samples taken over the subsequent 24h. The purpose was to assess the relative utility of the two types of samples. Twenty-five convalescent hospital in-patients were recruited as participants. Uranium concentrations in the 24-h samples ranged from 1 to 10.6 ng l(-1); in the spots, from not detectable to 38.1 ng l(-1). Normalised to creatinine, concentrations in the 24h samples ranged from approximately 100 to 800 ng mol(-1) creatinine; in the spot samples, from not detectable to approximately 4000 ng mol(-1) creatinine. The ranges appear similar to those reported for residents of the US. The distribution of spot sample results indicated that 95% of a participant's creatinine-adjusted concentrations from spot samples would be within the range 40-250% of his mean. Adjusting for creatinine almost entirely eliminated a slight indication of diurnal variation in urinary uranium concentration in spot samples. All the 24-h samples and 131 out of the 133 spot samples showed ratios of isotopes (238)U to (235)U consistent with natural uranium (i.e. neither enriched nor depleted). Slightly elevated ratios in two spot samples were not supported by other samples from the same participants, indicating that slightly elevated ratios may be recorded on very low concentration (<1 ng l(-1)) samples. In the main, quantification of this isotope ratio from spot samples was only slightly more variable than from 24-h samples. Complete 24-h urine samples gave better precision than spot samples in estimating uranium concentrations at these low levels, but presented more logistic difficulties in the collection of the samples. Clarification of the relative merits of alternative sampling strategies enables the design of a wider study to be optimised.