Ann. occup. Hyg., Vol. 48, No. 2, pp. 171-179, 2004
© 2004 British Occupational Hygiene Society
Published by Oxford University Press
Comparison of Uncertainties Related to Standardization of Urine Samples with Volume and Creatinine Concentration
1 National Institute of Occupational Health, Lersø Parkallé 105, DK-2100 Copenhagen; 2 Institute of Public Health, University of Copenhagen, c/o Depepartment of Pharmacology, Blegdamsvej 3, DK-2200 Copenhagen, Denmark
Received 20 May 2003; in final form 15 October 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
When measuring biomarkers in urine, volume (and time) or concentration of creatinine are both accepted methods of standardization for diuresis. Both types of standardization contribute uncertainty to the final result. The aim of the present paper was to compare the uncertainty introduced when using the two types of standardization on 24 h samples from healthy individuals. Estimates of uncertainties were based on results from the literature supplemented with data from our own studies. Only the difference in uncertainty related to the two standardization methods was evaluated. It was found that the uncertainty associated with creatinine standardization (19–35%) was higher than the uncertainty related to volume standardization (up to 10%, when not correcting for deviations from 24 h) for 24 h urine samples. However, volume standardization introduced an average bias of 4% due to missed volumes in population studies. When studying a single 24 h sample from one individual, there was a 15–20% risk that the sample was incomplete. In this case a bias of
25% was introduced when using volume standardization, whereas the uncertainty related to creatinine standardization was independent of the completeness of the sample. The uncertainty of creatinine standardization is increased when studying single voids rather than 24 h urine samples. This is partially counteracted by the increased statistical power due to the increased number of samples for each individual. Furthermore, there is a considerable increase in convenience for the participants, when collecting small volumes rather than complete 24 h samples.
Keywords: bias; standardization; uncertainty; urine
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The use of urinary biomarkers as exposure or effect measures is widespread in preventive and occupational health, e.g. exposure to polycylic aromatic hydrocarbons (Jongeneelen et al., 1988; Hansen et al., 1994), metals (Kraus et al., 2001) and the effects of occupational stress (Lundberg et al., 1994; Sluiter et al., 2000). Urine has the advantage that the individuals may collect it themselves, as no technical skills are required. Thus collection can be carried out at work, at home or elsewhere, interfering very little with the individual’s routine lifestyle.
In most cases excreted amounts of urinary biomarkers are used as indicators of the body burden of an exogenous (biomarker of exposure) or endogenous (biomarker of effect) substance. It can be obtained by measuring the concentration of the relevant biomarker, the total volume of the urine and the corresponding time interval, e.g. 24 h or time since last void. However, complete urine collection may be difficult to obtain, due to forgetfulness, misplacement of samples, lack of container capacity, erroneous inclusion of urine from the first void or loss of urine during defaecation (De Wardener, 1985). This is particularly so when studying individuals performing their routine lifestyles, as opposed to patients in, for example, a metabolic ward (Wheeler and Sheiner, 1979). Furthermore, it may be inconvenient to carry containers of urine around for a full day, which may reduce the number of individuals wanting to participate in a study. Thus an alternative is to collect smaller volumes and combine the biomarker measurement with a measure of diuresis, e.g. creatinine.
Both use of volume and use of creatinine concentration for standardization of diuresis contributes uncertainty to the final result. Furthermore, a bias is introduced for volume standardization due to incomplete collection (Knuiman et al., 1986; Harris et al., 2000). The uncertainty related to creatinine standardization is mainly related to the rate of excretion of creatinine. The creatinine excretion rate is higher for men than for women (Simpson et al., 1978), decreases with age (Simpson et al., 1978) and increases with exercise (Calles-Escandon et al., 1984), muscle mass (Edwards and Whyte, 1959) and intake of meat (Mayersohn et al., 1983; Laville et al., 1989; Herrera and Rodriguez-Iturbe, 1998). Furthermore, the excretion rate varies due to the diurnal (Pasternack and Kuhlback, 1971), seasonal (Ransil et al., 1977) and, for women, menstrual cycles (Davison and Noble, 1981; Paaby et al., 1987). These variations are all (except for seasonal variation) described by the within (CVi) and the between (CVg) individual variations. In both cases (volume and creatinine standardization), the uncertainty of the measurement will also introduce uncertainty in the final result. Hence, it is relevant to ask which measure, volume or creatinine standardization, is most accurate and under what circumstances.
The aim of the present paper was to compare the uncertainty that is introduced when using volume with the uncertainty introduced when using the concentration of creatinine, for standardization of diuresis in urine from healthy individuals. Only the difference in uncertainty related to the two standardization methods was evaluated. Uncertainties due to sampling, chemical analysis and variability within and between individuals were combined in uncertainty budgets for 24 h urine samples. Furthermore, relevant biases were discussed. Estimates of uncertainties were based on results from the literature supplemented with data from our own studies.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Uncertainty budget
In order to estimate the uncertainty associated with biomarker analysis in urine, the 24 h excretion of the biomarker (
B24 h) was defined as the quantity of interest, because this quantity can be related to the biomarker body burden under certain pharmacokinetic assumptions (Rappaport, 1993). In mathematical terms we have:
where [B]urine is the concentration of the biomarker in the 24 h urine, V is the volume and t is the length of the collection period (expressed in days of 24 h). t should ideally be one whole day. Hence, estimation of B24 h requires measurement of the concentration of the biomarker in urine and assessment of the length of the collection time and volume of the sample. The values of these quantities are all prone to uncertainty.
Alternatively, measurement of creatinine can be used to standardize for diuresis. The underlying assumption is that the excretion rate of creatinine is constant (Kcreatinine) both within and between individuals over time, i.e. creatinineurine/t = ([creatinine]urine·V)/t = Kcreatinine, where [creatinine]urine denotes the concentration of creatinine in urine. Combining the two expressions gives the following expression for the excretion of biomarker per day:
Note that Kcreatinine is expressed in mmol creatinine/day. Thus the excreted amount of biomarker per 24 h may alternatively be estimated by measurement of the concentration of the biomarker in urine combined with measurement of the creatinine concentration and multiplied by the creatinine excretion rate. Also, the values of these quantities are subject to uncertainty. Usually, results are expressed as amount of biomarker per amount of creatinine in the urine (for example, mmol ‘biomarker’/mmol creatinine). This corresponds to dividing equation 1B by the ‘constant’ excretion rate of creatinine (Kcreatinine), and the result is therefore a proxy measure for the 24 h excreted amount of biomarker under the assumption that Kcreatinine is constant. However, Kcreatinine is subject to biological variability, and in order to compare the two approaches (volume and creatinine standardization) with respect to uncertainty, we have chosen to maintain the expression in the form of 1B.
The uncertainty of the estimate of B24 h, expressed as relative standard deviation (CV) can be estimated by combining the CV values of the factors in equations (1A) and (1B) using the following formulae for volume and creatinine, respectively:
By comparing these two expressions, it can be quantitatively stated which of the two methods, volume or creatinine standardization, gives results with the least uncertainty. As can be seen, the uncertainty associated with the concentration of the biomarker in urine (CV[B]urine) appears in both expressions and therefore contributes equally to the uncertainty of the result regardless of the method used for standardization of diuresis. The difference between the uncertainty of the two methods is determined by the uncertainty related to the estimates of the excreted volume of urine (CVV), the corresponding time interval (CVt), the concentration of creatinine in urine ( 
) and the constancy of the excretion rate of creatinine from the body ( 
). Thus we will focus on these four terms. In order to achieve the best estimates, these were based on relevant data in the literature supplemented with data from our own studies, where available. The data from our own studies were based on secondary analysis of data collected for other purposes (Autrup et al., 1999; Knudsen et al., 1999; Loft et al., 1999). Where a single study in the literature gives a range of values these were combined to give a mean value by addition of variances. The extremes of the range of reported values for biological variation from the relevant studies have been used for calculation of the possible range of values for uncertainty related to creatinine standardization.
Study group
The study group was a sub-cohort of participants in the Danish Environmental Strategic Research Programme evaluating the impact of exposure to traffic-generated air pollution (Autrup et al., 1999; Knudsen et al., 1999; Loft et al., 1999). The study group comprised all bus drivers and mail carriers (n = 93), who had collected two 24 h urine samples (14 females and 79 males). Of the 186 24 h samples collected, 41 samples (22%) were reported to be incomplete. The mean self-estimated missed volume was 322 ml (see also Table 1). Thus 145 samples from 84 individuals (11 females and 73 males) were reported to be complete. The means (medians, ranges) of the censored sub-cohort (n = 84) were: age, 44 (45, 25–60) yr; height, 178 (179, 156–194) cm; weight, 84 (84, 40–115) kg; body mass index (BMI), 26.8 (26.5, 16.4–36.8) kg/m2.
|
All individuals signed an informed consent form and the local ethical committee approved the study protocol.
Study design
Matching values of creatinine concentration and urinary volume had been measured. Unfortunately, the actual duration of the 24 h collection period had not been noted. Estimates of biological variation were based only on the 145 samples reported to be complete. Creatinine excretion rate was estimated as concentration x volume/time and urine output rate as volume/time.
Chemical analysis
The creatinine method was based on the well-known Jaffe’s reaction (Bartels and Bohmer, 1971) and measured on a COBAS Mira autoanalyzer (Roche). The method characteristics have been described in detail elsewhere (Garde et al., 2003a).
Statistics
Statistical analysis was carried out by use of the SAS® SystemTM, version 8.02 (SAS Institute, Cary, NC). Due to non-normal (skewed) distributions and heteroscedastic variances, creatinine excretion rate was analysed on a logarithmic scale. Effects of gender, age, height, weight and urine output rate on creatinine excretion rate were analysed in repeated measures analysis of variance models with individual as a variance component with compound symmetry. Variables were successively omitted from the models if likelihood ratio tests produced probabilities >0.05. For estimation of within and between individual variation for creatinine excretion rate, data were analysed separately for women and men using a variance component model with individual as a variance component with compound symmetry. Variance components describing variations between individuals (Vg) and the total within individual variation (Vti), i.e. variation within individuals combined with analytical variation, were estimated. The latter includes unexplained variation. CV was calculated from the exact formulae CVg2 = eVg – 1 and CVti2 = eVti – 1. CVi was calculated as the square root of the difference between the total within individual variance (Vti) and the analytical variance. Anderson–Darlings test for normality was used to test the residuals from the variance component model.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Biological parameters estimated in the present study
The urinary volume was 2.0 [95% confidence interval (CI) 1.9–2.2] l per 24 h, the creatinine concentrations were 6.2 (CI 4.1–9.3) mmol/l for women and 9.5 (CI 8.5–10.5) mmol/l for men and the creatinine excretion rate was 12.3 (CI 10.5–14.5) mmol/24 h for women and 17.7 (CI 15.8–18.0) mmol/24 h for men. The 24 h creatinine excretion rate decreased with age [0.8% (CI 0.2–1.5%) per yr, P = 0.017] and was 40% (CI 20–70%) higher for men than for women (P < 0.001). Furthermore, there was a tendency (P = 0.09) for an association between the creatinine excretion rate and weight, whereas the urine output rate and height were not associated with the 24 h creatinine excretion rate in the present study. The CVi for the 24 h creatinine excretion rate without statistical control for age was 14% for women and 24% for men and CVg was 25% for women and 21% for men (see also Table 2).
|
Uncertainty budgets and bias
Uncertainty budgets, including estimates of the relevant uncertainties and biases for volume and creatinine standardization of 24 h urine samples, are presented in Table 3. The estimates used in the uncertainty budget for volume standardization included estimates of measurement uncertainty of volume and the length of time during which the samples were collected (Bingham and Cummings, 1985). The estimate of bias due to missed volume was based on studies using p-aminobenzoic acid (PABA) (Bingham and Cummings, 1985) (see Table 1). The estimates used in the uncertainty budget for creatinine standardization included uncertainty due to measurement (Garde et al., 2003a) and biological variation. The estimates of within and between individual variations of creatinine excretion rate were based on data presented in this study as well as in other studies with control of volume (Paterson, 1967; Wheeler and Sheiner, 1979; Bingham and Cummings, 1985; Harris et al., 2000) (see Table 2).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
When measuring biomarkers in urine, volume or concentration of creatinine are often used to standardize for diuresis. Both types of standardization contribute uncertainties to the final result. The purpose of the present paper was to compare the uncertainties and possible biases related to measurement of 24 h urine samples. Only the difference in uncertainty related to the two standardization methods was evaluated.
The main conclusions regarding the combined uncertainties, which were based on results from the literature supplemented with data presented here, were as follows:
- For both volume and creatinine standardization the uncertainties related to measurement, i.e. the estimates of volume and creatinine concentration, respectively, were relatively small.
- The combined uncertainty related to creatinine standardization (19–35%) was higher than the uncertainty related to volume standardization (up to 10%, when not correcting for deviations from 24 h).
- A bias towards lower values due to missed volume was introduced for volume standardization, but not creatinine standardization. The average bias was 4% in population studies and
25% when considering a single incomplete sample from an individual.
- The combined uncertainty related to creatinine standardization (19–35%) was higher than the uncertainty related to volume standardization (up to 10%, when not correcting for deviations from 24 h).
The uncertainty related to the collection period, i.e. deviations from 24 h, was unfortunately not determined in the present study and was reported in only a single published study. The mean time from the beginning to the end of 122 collections of 24 h urine samples was 23.5 h, with a SD of 2.2 h (9.4%) (Bingham and Cummings, 1985). This contributes considerably to the uncertainty of volume standardization, if not taken into consideration (Table 3). Hence, it is recommended that information on the actual period of collection is obtained in order to standardize to 24 h when using volume standardization. The uncertainty of measurement due to use of a measuring cylinder was relatively small (Table 3).
The biological variation (within and between individual variation) in the creatinine excretion rate constituted the main uncertainty component related to creatinine standardization. Both types of variations have been studied extensively for 24 h urine samples (Table 2). The measurement of creatinine excretion rate does by nature include the uncertainty of volume standardization, which must be minimized for the present purpose. Thus, only studies that controlled for incomplete samples were included in estimating the uncertainty component. The control was done either by use of PABA (Bingham and Cummings, 1985) or by self-report [data presented in this study together with studies by Paterson (1967), Wheeler and Sheiner (1979) and Harris et al. (2000)]. Self-reports were included although there was a risk of overestimating the uncertainty due to samples with unreported missed volume. None of the studies included information on correction for the actual length of the 24 h period. Furthermore, the focus was on studies with relatively heterogeneous study groups in order to reflect the true biological variation, which might be artificially small in homogeneous study groups.
Estimates of the within-individual variation in the creatinine excretion rate are given in Table 2. While our estimate of CVi for women (14%) is in the same range as that of others, our estimate for men (24%) is higher than other estimates based on studies with control of volume. There is no a priori reason why the male creatinine excretion rate should exhibit a greater within-individual variation than the female creatinine excretion rate. The highest within-individual variation was found in studies without control of completeness of samples (Wheeler and Sheiner, 1979; Shephard et al., 1981), which was expected, since the variation in the creatinine excretion rate will be biased towards higher values if missed volumes were not accounted for. For some studies without control of volume (Bailey and De Wardener, 1970; Greenblatt et al., 1976; Gowans and Fraser, 1988) the reported within-individual variations were comparable to the estimate found in studies with control of volume (Paterson, 1967; Wheeler and Sheiner, 1979; Harris et al., 2000). The nature of the study group (laboratory staff and hospitalized patients) in these studies made it likely that the participants had adhered to the collection protocol more carefully than ‘random’ study populations. In other studies ‘outliers’ based on statistical evaluations had been excluded before estimating the within-individual variation (Curtis and Fogel, 1970; Ricós et al., 1994), thereby introducing a bias towards smaller estimates. In studies where missed volumes were controlled for and outliers were not excluded on a statistical basis, the range of within-subject variation of creatinine excretion rate during 24 h was 9–24% (Paterson, 1967; Wheeler and Sheiner, 1979; Bingham and Cummings, 1985; Harris et al., 2000).
The creatinine excretion rate decreases with age and is found to be higher for men than for women (Simpson et al., 1978), which was confirmed in the present study. Since creatinine excretion is related to lean body mass (Edwards and Whyte, 1959), a likely explanation is the lower fat-free mass of women compared to men (Hughes et al., 2002) and a possible decrease in the muscle mass with age (Forbes, 1999; Guo et al., 1999; Hughes et al., 2002) and therefore lower turnover of creatine and creatine phosphate to creatinine. Estimates of the between individual variation in the creatinine excretion rate are presented in Table 2. The estimates of CVg in the only study found in the literature where incomplete samples were controlled for were 18% for men and 17% women (Bingham and Cummings, 1985). This corresponded well with the estimates found in the present study (21 and 25% for men and women, respectively). In studies not controlling for incomplete samples the range of CVg was 6–36% (Shephard et al., 1981; Gowans and Fraser, 1988; Ricós et al., 1994; Newman et al., 2000). The differences between studies might be explained by the group sizes, degree of homogeneity of the groups and the method for control of volume collection. The uncertainty may be reduced further if the groups are homogeneous with respect to age. For the present purpose, estimates were based on studies with a heterogeneous study group and control of volume collection (Bingham and Cummings, 1985; data presented here), rendering a range of estimates of 17–25% for the uncertainty related to the between-individual variation of the creatinine excretion rate.
When the uncertainties related to standardization of the 24 h urine samples were combined in uncertainty budgets (Table 3), the uncertainty related to volume standardization (up to 10%) was lower than the uncertainty related to creatinine standardization (19–35%). The latter uncertainty may be reduced somewhat by controlling for factors which affect the excretion rate of creatinine, e.g. age, strenuous physical activity and, in particular, meat intake. Alternatively, the number of participants may be increased compared to a study using volume standardization in order to achieve the necessary statistical power.
The uncertainty of the excretion rate of the biomarker of interest is not included in the uncertainty budgets. The elimination constant is related to the mechanism of excretion in the kidneys, and is important when urinary measurements are used to estimate the body burden of the biomarker. However, the uncertainty component related to it is the same regardless of the applied standardization method (creatinine or volume). Thus creatinine standardization is as applicable as volume standardization regardless of the excretion mechanism of the biomarker of interest (and whether this mechanism is known). In the case of the uncertainty of the excretion rate of the biomarker of interest being very large, the impact of the difference between the uncertainties related to creatinine and volume standardization could even be negligible.
Apart from deliberate dilution of the urine, which may be relevant in drug testing, all possible mistakes in collection of urine samples (completely or partially missed voids) result in smaller volumes. Thus the estimate of urine volume introduces a bias rather than an uncertainty. The bias towards smaller volumes means that the total dose will be underestimated when using volume standardization. The number of incomplete 24 h samples has been estimated in a few studies (Table 1), where the self-reported number of incomplete samples ranged from 3 to 18% (Bingham and Cummings, 1985; Knuiman et al., 1986; Harris et al., 2000). Compared with this, the percentage of self-reported incomplete samples (22%) in the present study is at the higher end. Differences between countries (Knuiman et al., 1986) may offer one explanation. Another reason could be differences between the study protocols and the study groups of interest, e.g. their commitment and the possibility to urinate at fixed places, thereby reducing the need for carrying bottles around. The study group in the present work contained bus drivers and mail carriers, who did not have easy access to sanitary facilities when ‘on route’, thereby making it more difficult to obtain complete samples despite a high degree of commitment. Furthermore, a study using recovery of PABA indicated that the estimate of incomplete 24 h samples based on self-reports might be in the lower range. Based on 24 h samples from participants who reported having ingested the full dose of PABA, an estimate of 17% incomplete samples was obtained (Bingham and Cummings, 1985). In corresponding studies a larger range of missed volume (0–47%) was estimated based on the recovery of ingested PABA (Bingham and Cummings, 1985; Knuiman et al., 1986). However, these studies did not control for the intake of PABA. Hence, it could not be excluded that the low recovery of PABA might be due to a low intake of PABA (Knuiman et al., 1986). Based on the available information, it seemed reasonable to assume that the typical percentage of incomplete 24 h samples in non-hospitalized individuals was 15–20%.
The average missed volume in an incomplete 24 h sample based on self-reports found in the present study was comparable to that found in other studies of non-hospitalized individuals (Table 1). However, a missed volume might be very difficult to estimate. Therefore, the estimate used in the present uncertainty budget was based on a study using recovery of PABA. The recovery of PABA in 24 h samples judged as incomplete was 69 and 73% for men and women, respectively, corresponding to missed volumes of 31 and 27%, respectively (Bingham and Cummings, 1985). Although all participants reported to have ingested PABA as required, this estimate may be at the higher end. Hence, an estimate of 25% missed volume was adopted for incomplete 24 h samples. In the same study the missed volume was 4% of the total collected volume, which corresponds well with the above estimates of incomplete samples and missed volumes per sample. Thus the average bias introduced by incomplete samples was 4% in population studies, but increased to 25% when considering an incomplete 24 h sample from a single individual. In contrast, incomplete samples introduced no bias when using creatinine standardization. Thus, in the case of a (possibly) incomplete sample, creatinine standardization may be preferred over volume standardization.
So far, this paper has focused on the average daily excretion of a biomarker as measured in 24 h urine samples. However, there is a growing interest in studying the variability in or profile of excretion of biomarkers based on single voids, e.g. over a day (Sluiter et al., 2000; Garde et al., 2003b) and it is therefore relevant to ask if the results from 24 h urine samples may be transferred to single voids. Preliminary results from our own data indicate that the within and between individual variation in creatinine excretion increases when using single voids compared to 24 h urine samples. This may be explained by a circadian variation in the creatinine excretion rate (Pasternack and Kuhlback, 1971) and an acute increase in the excretion rate after intake of boiled meat (Mayersohn et al., 1983; Herrera and Rodriguez-Iturbe, 1998). The average missed volume is expected to be the same, as are the uncertainties of the chemical analysis. Thus the uncertainty of creatinine standardization of a single void is further increased compared to 24 h urine samples. However, this is to some degree counteracted by the increased statistical power due to more samples per individual. Furthermore, samples for creatinine standardization are much more convenient to collect in a daily setting because they require collection of only a small volume of the void and the time of sampling. In contrast to volume standardization, it is not necessary to know the timespan, and therefore the time of the last void is irrelevant. Lastly, it should be noted that biologically threshold limits established from a representative population will include the biological intra- and inter-individual variation in creatinine excretion rate. Hence, the average health practitioner, who compares results from single voids, e.g. end of shift void, with biological threshold limits need only care about the uncertainty related to the chemical analysis.
In conclusion, volume standardization has a lower uncertainty than creatinine standardization. However, an average bias of 4% is introduced when using volume standardization in population studies. This bias is increased considerably (to 25%) when studying a single, incomplete sample. The uncertainty of creatinine standardization is increased when studying single voids. This is partially counteracted by the increased statistical power due to the increased number of samples for each individual. Furthermore, there is a considerable increase in convenience for the participants when collecting small volumes rather than complete 24 h samples.
Acknowledgement—The study was supported by the Danish Environmental Research Programme.
|
FOOTNOTES |
|---|
* Author to whom correspondence should be addressed. Tel: +45-3916-5258; fax: +45-3916-5201; e-mail: ahg{at}ami.dk
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Autrup H, Daneshvar B, Dragsted LO et al. (1999) Biomarkers for exposure to ambient air pollution—comparison of carcinogen–DNA adduct levels with other exposure markers and markers for oxidative stress. Environ Health Perspect; 107: 233–8.
Bailey RR, De Wardener HE. (1970) Creatinine excretion. Lancet; 1: 145.[CrossRef][Web of Science][Medline]
Bartels H, Bohmer M. (1971) Eine mikrometode zur kreatininibestimmung. Clin Chim Acta; 32: 81–5.[CrossRef][Web of Science][Medline]
Bingham SA, Cummings JH. (1985) The use of creatinine output as a check on the completeness of 24-hour urine collections. Hum Nutr Clin Nutr; 39C: 343–53.
Calles-Escandon J, Cunningham JJ, Snyder P et al. (1984) Influence of exercise on urea, creatinine, and 3-methylhistidine excretion in normal human subjects. Am J Physiol; 246: E334–8.
Curtis G, Fogel M. (1970) Creatinine excretion: diurnal variation and variability of whole and part-day measures. A methodologic issue in psychoendocrine research. Psychosom Med; 32: 337–50.
Davison JM, Noble MC. (1981) Serial changes in 24 hour creatinine clearance during normal menstrual cycles and the first trimester of pregnancy. Br J Obstet Gynaecol; 88: 10–7.[Web of Science][Medline]
De Wardener HE. (1985) Tests of glomerular functional integrity and proteinuria. In De Wardener HE, editor. The kidney. Edinburgh: Churchill Livingstone: 39.
Edwards KDG, Whyte HM. (1959) Creatinine excretion and body composition. Clin Sci; 18: 361–6.[Web of Science][Medline]
Edwards OM, Bayliss RI, Millen S. (1969) Urinary creatinine excretion as an index of the completeness of 24-hour urine collections. Lancet; 2: 1165–6.[Web of Science][Medline]
Forbes GB. (1999) Longitudinal changes in adult fat-free mass: influence of body weight. Am J Clin Nutr; 70: 1025–31.
Garde AH, Hansen ÅM, Kristiansen J. (2003a) Evaluation, inclding effects of storage and repeated freeezing and thawing, of a method for measurement of urinary creatinine. Scand J Clin Lab Invest; in press.
Garde AH, Hansen ÅM, Persson R, Ohlsson K, Ørbæk P. (2003b) The influence of production systems on physiological responses measured in urine and saliva. Stress Health; 19: 297–306.[CrossRef]
Gowans EM, Fraser CG. (1988) Biological variation of serum and urine creatinine and creatinine clearance: ramifications for interpretation of results and patient care. Ann Clin Biochem; 25: 259–63.
Greenblatt DJ, Ransil BJ, Harmatz JS, Smith TW, Duhme DW, Koch-Weser J. (1976) Variability of 24-hour urinary creatinine excretion by normal subjects. J Clin Pharmacol; 16: 321–8.[Abstract]
Guo SS, Zeller C, Chumlea WC, Siervogel RM. (1999) Aging, body composition, and lifestyle: the Fels Longitudinal Study. Am J Clin Nutr; 70: 405–11.
Hansen ÅM, Omland Ø, Poulsen OM et al. (1994) Correlation between work process-related exposure to polycyclic aromatic hydrocarbons and urinary levels of 
-naphthol, ß-naphthylamine and 1-hydroxypyrene in ironfoundry workers. Int Arch Occup Environ Health; 65: 385–94.[CrossRef][Web of Science][Medline]
Harris SA, Purdham JT, Corey PN, Sass-Kortsak AM. (2000) An evaluation of 24-hour urinary creatinine excretion for use in identification of incomplete urine collections and adjustment of absorbed dose of pesticides. Am Ind Hyg Assoc J; 61: 649–57.
Herrera J, Rodriguez-Iturbe B. (1998) Stimulation of tubular secretion of creatinine in health and in conditions associated with reduced nephron mass. Evidence for a tubular functional reserve. Nephrol Dial Transplant; 13: 623–9.
Hughes VA, Frontera WR, Roubenoff R, Evans WJ, Singh MA. (2002) Longitudinal changes in body composition in older men and women: role of body weight change and physical activity. Am J Clin Nutr; 76: 473–81.
Jackson S. (1966) Creatinine in urine as an index of urinary excretion rate. Health Phys; 12: 843–50.[Web of Science][Medline]
Jongeneelen FJ, Bos RP, Henderson PT. (1988) Metabolites of PAH in urine of exposed workers. Toxicol Environ Chem; 16: 295–307.
Knudsen LE, Norppa H, Gamborg MO et al. (1999) Chromosomal aberrations in humans induced by urban air pollution: influence of DNA repair and polymorphisms of glutathione S-transferase M1 and N-acetyltransferase 2. Cancer Epidemiol Biomarkers Prev; 8: 303–10.
Knuiman JT, Hautvast JG, van der Heijden L et al. (1986) A multi-centre study on completeness of urine collection in 11 European centres. I. Some problems with the use of creatinine and 4-aminobenzoic acid as markers of the completeness of collection. Hum Nutr Clin Nutr; 40: 229–37.[Medline]
Kraus T, Schramel P, Schaller KH, Zobelein P, Weber A, Angerer J. (2001) Exposure assessment in the hard metal manufacturing industry with special regard to tungsten and its compounds. Occup Environ Med; 58: 631–4.
Laville M, Hadj-Aissa A, Pozet N, Le Bras JH, Labeeuw M, Zech P. (1989) Restrictions on use of creatinine clearance for measurement of renal functional reserve. Nephron; 51: 233–6.[Web of Science][Medline]
Loft S, Poulsen HE, Vistisen K, Knudsen LE. (1999) Increased urinary excretion of 8-oxo-2'-deoxyguanosine, a biomarker of oxidative DNA damage, in urban bus drivers. Mutat Res; 441: 11–9.[Web of Science][Medline]
Lundberg U, Mårdberg B, Frankenhaeuser M. (1994) The total workload of male and female white collar workers as related to age, occupational level, and number of children. Scand J Psychol; 35: 315–27.[Web of Science][Medline]
Mayersohn M, Conrad KA, Achari R. (1983) The influence of a cooked meat meal on creatinine plasma concentration and creatinine clearance. Br J Clin Pharmacol; 15: 227–30.[Web of Science][Medline]
Newman DJ, Pugia MJ, Lott JA, Wallace JF, Hiar AM. (2000) Urinary protein and albumin excretion corrected by creatinine and specific gravity. Clin Chim Acta; 294: 139–55.[CrossRef][Web of Science][Medline]
Paaby P, Brochner-Mortensen J, Fjeldborg P, Raffn K, Larsen CE, Moller-Petersen J. (1987) Endogenous overnight creatinine clearance compared with 51Cr-EDTA clearance during the menstrual cycle. Acta Med Scand; 222: 281–4.[Web of Science][Medline]
Pasternack A, Kuhlback B. (1971) Diurnal variations of serum and urine creatine and creatinine. Scand J Clin Lab Invest; 27: 1–7.[Web of Science][Medline]
Paterson N. (1967) Relative constancy of 24-hour urine volume and 24-hour creatinine output. Clin Chim Acta; 18: 57–8.[CrossRef][Web of Science][Medline]
Ransil BJ, Greenblatt DJ, Koch-Weser J. (1977) Evidence for systematic temporal variation in 24-hour urinary creatinine excretion. J Clin Pharmacol; 17: 108–19.[Abstract]
Rappaport SM. (1993) Biological considerations in assessing exposures to genotoxic and carcinogenic agents. Int Arch Occup Environ Health; 65: S29–35.[CrossRef][Web of Science][Medline]
Ricós C, Jiménez CV, Hernández A et al. (1994) Biological variation in urine samples used for analyte measurements. Clin Chem; 40: 472–7.
Shephard MD, Penberthy LA, Fraser CG. (1981) Short- and long-term biological variation in analytes in urine of apparently healthy individuals. Clin Chem; 27: 569–73.
Simpson FO, Nye ER, Bolli P et al. (1978) The Milton Survey: Part 1, General methods, height, weight and 24-hour excretion of sodium, potassium, calcium, magnesium and creatinine. N Z Med J; 87: 379–82.[Web of Science][Medline]
Sluiter JK, Frings-Dresen MH, van der Beek AJ, Meijman TF, Heisterkamp SH. (2000) Neuroendocrine reactivity and recovery from work with different physical and mental demands. Scand J Work Environ Health; 26: 306–16.[Web of Science][Medline]
Wheeler LA, Sheiner LB. (1979) Clinical estimation of creatinine clearance. Am J Clin Pathol; 72: 27–32.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  E. C. Bruce, Y. Guo, K. C. Lawson, A. K. Manatunga, S. F. Auyeung, W. M. McDonald, N. Rushing, A. R. Brown, N. Gilles, M. Emery, et al. Platelet Thromboxane A2 Secretion in Patients With Major Depression Responsive to Electroconvulsive Therapy Psychosom Med, April 1, 2008; 70(3): 319 - 327. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Remer, T. Dimitriou, and C. Maser-Gluth Renal Net Acid Excretion and Plasma Leptin Are Associated with Potentially Bioactive Free Glucocorticoids in Healthy Lean Women J. Nutr., February 1, 2008; 138(2): 426S - 430S. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||