Urine trace element concentrations

As discussed before, quadrupole based ICP-MS allows multi-element determination at the trace and ultratrace level and/or isotope ratios in aqueous solutions in a few minutes as a routine method with detection limits of elements in the sub pgml-1 range and a precision for determined trace element concentration in the low % range (RSD - relative standard deviation). The precision for isotope ratio measurements varies between 0.1% and 0.5% RSD. This isotope ratio precision is sufficient for a multitude of applications, e.g., for evidence of contamination of sample with depleted or enriched uranium in urine (this technique is used in the author s laboratory in a routine mode14) or the isotope dilution technique for the quantitative determination of trace element and species concentration after doping the sample with enriched isotope spikes. 

Table II. Comparison of Trace Element Concentrations for Human Urine with ICP—AES Limits of Quantitative Determination...

The state of the art with respect to the potential of ICP-AES for the determination of trace elements in urine can be assessed by comparing the model concentration and range of concentrations for each of the elements (computed from the data in Table I) to the corresponding experimental limit of quantitative determination (LQD) for that element in aqueous solutions with 1% (vol) nitric acid, as is done in Table II. If the trace element concentrations in urine are as shovm in the table, and if the LQDs which can be achieved for urine samples are not substantially different from those observed for 1% nitric acid solutions, then ICP-AES should be applicable for the quantitative determination of many of the trace elements occurring in urine. 

The important features of the sample preparation procedure were as follows. First, the samples were acidified to dissolve normal urine precipitates and to prevent analyte loss by adsorption on the walls of the sample containers (13). Second, the procedure was kept as simple as possible so that the risk of contamination and/or loss was minimized. Third, dilute, normal, and concentrated series of solutions were used to simulate actual urine samples with a wide range of total dissolved solids. Fourth, because the rate of sample nebulization and the corresponding rate of sample introduction into the plasma can be aflFected by changes in the amount of total dissolved solids, internal reference elements were included in each sample and reference solution. The use of analyte/internal reference element net intensity ratios provided a means of correcting for possible diflFerences in sample introduction rate according to the internal reference principle (14,15). Finally, because all of the sample solutions introduced into the plasma were derived from one composite, the different series were known to have trace element concentrations which were related to each other by known dilution factors (see Table IV). 

The analytical results provided few surprises with respect to the trace element concentrations which were expected for urine. See Tables II and V. The results for all of the elements studied were within or below the range of previously reported concentrations. The selenium and zinc results were shghtly greater than the corresponding model concentrations and those for aluminum, beryllium, cadmium, cobalt, chromium, iron, and vanadium were approximately one order of magnitude lower than the model values. Because of possible diflFerences in diet and because the samples studied in this work were derived from first-morning voids, it is not possible to draw unequivocal conclusions concerning the diflFerences between the observed and expected results. 

The benefit of urine analysis in clinical chemistry and occupational health medicine for diagnosis and therapy control is indisputable. Plasma emission spectrometry can deliver a lot of possibly important information about trace element concentrations, which can not be obtained simultaneously by any other analytical technique (Schramel et al., 1985). 

Today the preferred method of hair analysis is AAS on account of its sensitivity, validity, and economical nature. Other methods such as inductively coupled plasma photometry (ICP) [55], X-ray fluorescence [48,56], and activation analysis [57] make multielement analysis possible. Element determination in human hair is not as difficult as in blood and urine because the trace element concentrations are several powers of 10 higher, so that the matrix effect is not so important. 

Biomonitoring of environmental and occupationally relevant trace and ultratrace metals (Al, Co, Cr, Cu, Fe, Mn, Ni, Pt, V and Zn) in human serum and urine was carried out using ICP-SFMS at different mass resolutions by Begerow et cd 41 Whereas the elements free from isobaric interferences (Cd, Mn, Pb, Pt and Tl) were measured at low mass resolution (ml Am = 300), the determination of Al, Co, Cr, Cu, Fe, Ni, V and Zn was performed in the medium mass resolution mode (m/Am = 3000).41 Trace metal concentrations (Al, Ba, Be, Bi, Cd, Co, Cr, Hg, Li, Mn, Mo, Ni, Pb, Sb, Sn, Sr, Tl, V, W and Zr) in serum and blood samples from patients with Alzheimer s disease and healthy individuals measured by ICP-SFMS were compared by et al42 An increment of Hg and Sn in serum, higher levels of Co, Li, Mn and Sn and lower levels of Mo in blood were found in Alzheimer s disease samples.42... 

Such is not the case with zinc deficiency. While symptoms of zinc deficiency can be as vague as those in iron deficiency, there is no simple diagnostic test which defines the condition. Zinc concentrations in blood, urine or hair commonly do not accurately reflect body zinc status (1 -3) Although zinc is the major trace element found in fixed body tissues less than 1% of the approximate 2.5 g of total body zinc is circulating in blood ( ) while for iron, the major trace element found in the circulation, over 50% of the approximate 4.0 g of total body iron is found in blood and blood forming systems (5). 

Inductively coupled plasma-atomic emission spectrometry was investigated for simultaneous multielement determinations in human urine. Emission intensities of constant, added amounts of internal reference elements were used to compensate for variations in nebulization efficiency. Spectral background and stray-light contributions were measured, and their effects were eliminated with a minicomputer-con-trolled background correction scheme. Analyte concentrations were determined by the method of additions and by reference to analytical calibration curves. Internal reference and background correction techniques provided significant improvements in accuracy. However, with the simple sample preparation procedure that was used, lack of sufficient detecting power prevented quantitative determination of normal levels of many trace elements in urine. 

For certain sample types, both stray light and recombination eflFects on the analytical results can be largely eliminated by matrix matching. Such an approach is not practical for the determination of trace elements in urine because of the wide ranges of calcium and magnesium concentrations which occur in so-called normal samples. 

Table V summarizes the quantitative results obtained for 13 trace elements in urine. Both the results from the method of additions and those obtained from the 1% NaCl analytical cahbration curves are given. Concentrations were determined for dilute, normal, and concentrated urine solutions but, for ease of comparison, the results listed in the table are all reported in terms of the concentration of analyte present in the original composite urine sample. Background correction was performed for each sample and reference solution according to the wavelengthprofiling procedure outlined above. A 10-second photocurrent measure-...

In clinical analysis, flame AAS is very useful for serum analysis. Ca and Mg can be determined directly in serum samples after a 1 50 dilution, even with microaliquots of 20-50 pL [314]. In the case of Ca, La3+ or Sr2+ are added so as to avoid phosphate interferences. Na and K are usually determined in the flame emission mode, which can be realized with almost any flame AAS instrument. The burner head is often turned to shorten the optical path so as to avoid self-reversal. For the direct determination of Fe, Zn and Cu, flame AAS can also be used but with a lower sample dilution. Determination of trace elements such as Al, Cr, Co, Mo and V with flame AAS often requires a pre-concentration stage, but in serum and other body fluids as well as in various other biological matrices some of these elements can be determined directly with furnace AAS. This also applies to toxic elements such as Ni, Cd and Pb, which often must be determined when screening for work place exposure. When aiming towards the direct determination of the latter elements in blood, urine or serum, matrix modification has found wide acceptance in working practices that are now legally accepted for work place surveillance, etc. This applies e.g. for the determination of Pb in whole blood [315] as well as for the determination of Ni in urine (see e.g. Ref. [316]). 

Analytical Methods for Urine and Blood. Specific biomarkers of lewisite exposure are currently based on a very limited number of in vitro experiments (Jakubowski et al., 1993 Wooten et al., 2002) and animal studies (Logan et al., 1999 Fidder et al., 2000). Wooten et al. (2002) developed a solid-phase microextraction (SPME) headspace sampling method for urine samples followed by GC-MS analysis. It is the most sensitive method reported to date with a lower limit of detection of 7.4 pg/mL. Animal experiments have been limited in number and in their scope. In one study of four animals, guinea pigs were given a subcutaneous dose of lewisite (0.5 mg/kg). Urine samples were analyzed for CVAA using both GC-MS and GC coupled with an atomic emission spectrometer set for elemental arsenic (Logan et al., 1999). The excretion profile indicated a very rapid elimination of CVAA in the urine. The mean concentrations detected were 3.5 pg/mL, 250 ng/mL, and 50 ng/mL for the 0-8, 8-16, and 16-24 h samples, respectively. Trace level concentrations... 

When trace elements are analyzed in living subjects, the specimens that are available are limited. Usually only blood, urine, faeces or hair are available. Thus the tissues or organs of most interest, e.g. target organs for a toxic effect, may not be directly sampled and analyzed. The physiological factors that affect the relationship between the concentrations of a trace element in the target tissues and in the body fluids are thus very important. 

Usually, only spot urine specimens are available for trace element analysis. Because the concentration of many analytes is dependent on the rate of urine excretion, which varies to a great extent even in healthy people (Shephard et al., 1981 Young, 1979), some standardisation for urinary excretion rate has long been used in the assessment of exposure to toxic elements (Levine and Fahy, 1945 Molyneux, 1966 Elkins et ai., 1974). The most widely used approaches have been based on relative density, the concentration of creatinine in urine, and the length of the urine collection period, i.e. excretion rate. Araki and co-workers have extensively studied the correction of urine concentrations to a standard urinary flow rate of 1 mL/min in circumstance, where the water intake has been changed (Araki, 1980 Araki and Aono, 1989 Araki et al.. 1990). Although this approach cannot be applied in routine trace element analysis and is not necessarily representative of other situations where the renal treatment of trace elements and water varies, it provides a useful method to compare the behaviour of the excretion of different chemicals in the urine by the following equation ... 

Especially toxic concentrations of trace elements are not constant, but change with time, and often show an exposure-related fluctuation. For example, the half-time of nickel and chromium in exposure to water soluble compounds in the urine is 1-2 days (Tossa-vainen et al., 1980). Thus, to be able to interpret the concentrations, the time since exposure must be standardized. 

Hydrochloric and nitric acids have been used as preservatives for urine specimens for metal analyses, and mineral acids are extensively used in graphite furnace analysis of trace elements in biological specimens (e.g. Gills et al., 1974 Stoeppler and Brandt, 1980). Historically, concentrations of trace elements in commercially available acids have been incompatible with analysis of trace elements in biological samples (Kuehner et al., 1972). However, present commercial ultra pure hydrochloric, nitric, sulphuric and perchloric acids have been reported to be suitable for trace element analysis in urine without further purification (Golimowski et al., 1979 Brown et al., 1981 Veillon et al., 1982). 

For significant estimation of such analytical results, it is necessary to take into account only the total excretion per day of trace elements. The concentrations of trace elements in spontaneously taken urine-samples are strongly dependent on the nutrition behaviour of the probands (Fig. 18). 

CONCENTRATIONS OF MINERAL AND TRACE ELEMENTS IN HUMAN URINE SAMPLES... 

---