LAUSR

BACKGROUND Urine is often used for biomonitoring the exposure to elements. However, most studies report concentrations in spot urine samples, which may not accurately mirror the "gold standard" of complete 24-h (24 h) urine samples. There are relatively few data published for 24 h samples, and little information on the within- and between person variability. OBJECTIVES The present study aimed at assessing variability within and between individuals in 24 h excretion for a number of elements in adults from the general population and the typical 24 h excretion of these elements. In addition, we assessed concentrations adjusted for creatinine and specific gravity (SG), and associations between elements. METHODS 60 healthy non-smokers (31 women and 29 men) from Sweden, aged 21-64 years, collected all urine during 24 h (split into six separate samples) on two occasions, about one week apart. Concentrations of As, Br, Cd, Co, Cr, Cu, Fe, Hg, Li, Mn, Mo, Ni, P, Pb, S, Sb, Se, Sn, U, V, W, and Zn in urine were analyzed by inductively coupled plasma sector-field mass spectrometry (ICP-SF-MS) and 24 h excretion rates were calculated for each day. The ratio of between-individual variance and the total variance, the intra-class correlation (ICC) was calculated based on natural log-transformed 24 h excretion. Correlation coefficients were calculated between excretion rates (mass/24 h), and concentrations adjusted for creatinine and SG. RESULTS Geometric means (GM), and 90-percentiles are presented for each element. The 24 h excretion was higher in men than in women for most elements, and the difference was statistically significant for Cr, Cu, Fe, Li, P, Pb, S, Se, U, V, and Zn. However, for Cd and Co, the excretion was higher in women. Variability between days was low for Cd, Co, Hg, Pb, Sn, Se, V, and Zn (ICC 0.75-0.90), highest for Cr (ICC = 0.3) and Sb (ICC = 0.18), and moderate for the other elements. Spearman's rank correlation coefficients were about 0.8-0.9 for 17 elements, and 0.3-0.7 for Br, Cu, P, S, Se. Excretion of P and S were highly correlated, and also associated with excretion of most of the other elements, especially Cu, Se, V, and Zn. A high correlation was also found between As and Hg, between Mo and W, as well as between Cr, Fe and Mn. CONCLUSIONS These data present normal variability of 24 h excretion of a number of elements, and can also be used as updated reference levels for elements with no or limited previous literature available. Information on variability within- and between individuals is important to know when designing studies with urine levels of elements used as exposure biomarker in studies of associations with health outcomes.