Health Stream Literature Summary - Issue 54 - June 2009

Nephrotoxicity of uranium in drinking water from private drilled wells.
Seldén, A.I., Lundholm, C., Edlund, B., Högdahl, C., Ek, B.M., Bergström, B.E. and Ohlson, C.G. Environmental Research, (2009) 109(4): 486-494.

Uranium is commonly found in bedrock in Sweden, and levels exceeding 15 micro g/l have been found in 23-40% of water samples from drilled private drinking water wells in several parts of the country. This paper reports an investigation of kidney function and uranium exposure from drinking water in residents of the Arjang municipality, Varmland county in western Sweden.

The study population consisted of individuals 18-74 years of age and included permanent residents in a rural part of Arjang municipality with drilled wells (exposed group), and others from the central Arjang community with a municipal water supply (control group). Exposed subjects and controls were mailed a questionnaire covering their drinking water consumption, health, use of prescribed pharmaceuticals, dietary supplements and tobacco, work exposures and leisure-time activities. Information was also requested on the characteristics of the well they used, including any water purification equipment. Those in the exposed group were requested to collect two water samples: one for radon analysis and one for analysis of selected metals (cadmium, lead, mercury and uranium). They were also asked to provide a morning urine specimen for biochemical analysis. Controls were asked to provide a morning urine sample but municipal water samples were only requested from a small number of controls. Urine samples were tested for uranium, lead and biochemical markers of kidney function. Urine and water tests were conducted without knowledge of the exposure/control status of the participant.

There were 301 individuals representing 156 households in the exposed group and 152 individuals representing 96 households in the control group. Of the 153 well water samples available for analysis, about one-third (32.7%) contained at least 15 micro g/l of uranium, the recommended Swedish action level and WHO provisional guideline value. There were 8 samples which contained 100 micro g/l uranium or more, and the maximum detected level was 470 micro g/l. The median and mean uranium levels in the well water samples were 6.7 and 25.2 micro g/l, respectively, which was considerably higher than the levels in all of the control water samples (less than 0.2 micro g/l). One well water sample had a high lead level but no water samples had elevated levels of cadmium or mercury. There was a fairly weak but highly statistically significant correlation (r = 0.39; p less than 0.001) between uranium and radon levels in the well water samples which increased (to r = 0.48) after excluding wells with water quality improvement equipment.

The geometric mean uranium level in the exposed group's urine samples was more than eight times that of the controls (38 ng/l vs. 4.3 ng/l). There was a strong curvilinear correlation (r2 = 0.66; p less than 0.001) found between uranium levels in the water samples and urine from corresponding subjects. There was no statistically significant difference between the two groups for 7 out of 9 markers of kidney function. For other two markers (calcium and the enzyme N-acetyl-beta-D-glucosaminidase) mean levels were significantly higher in the control group, but these differences disappeared when adjusted for background variables (age, gender and smoking). However when exposure was defined in terms of uranium levels in the subjects' urine samples (without regard to exposed or control status), statistically significant correlations were found between the exposure and the markers beta2-microglobulin, kappa chains and protein HC, but no clear dose-response was found between the two highest exposure categories. When subjects with diabetes were excluded, a tendency towards a dose-response was seen for beta2-microglobulin, kappa chains and protein HC, suggesting a uranium-associated nephrotoxicity.

Uranium in urine was found to be strongly related to uranium levels in drinking water from drilled wells. Overall there were no significant differences found in kidney morbidity or function between exposed subjects and controls. However after taking into account the entire range of uranium exposure and several background variables, including diabetes, a weak exposure-related effect was seen in urinary levels of several kidney function biomarkers. The clinical relevance of these findings is still uncertain.

Comment: Naturally occurring uranium has chemical toxicity properties which affect kidney function at concentration below those where radioactivity would have an adverse effect. Guideline values are set on the basis of preventing chemical toxicity and thus provide an even higher degree of safety in terms of radiological exposure.

© Copyright Water Quality Research Australia Limited http://www.wqra.com.au/
Health Stream articles may be reproduced and communicated to third parties provided WQRA is acknowledged as the source. Literature summaries are derived in part from copyright material by a range of publishers. Original sources should be consulted and acknowledged.