Skip Navigation


NDT Advance Access originally published online on November 6, 2007
Nephrology Dialysis Transplantation 2008 23(4):1252-1256; doi:10.1093/ndt/gfm729
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
23/4/1252    most recent
gfm729v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Andersson, L.
Right arrow Articles by Barregard, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Andersson, L.
Right arrow Articles by Barregard, L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author [2007]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org


Methodological issues on the use of urinary alpha-1-microglobuline in epidemiological studies

Lena Andersson1, Börje Haraldsson2, Caroline Johansson1 and Lars Barregard1

1 Department of Occupational and Environmental Medicine and Sahlgrenska University Hospital and Academy, Göteborg University, Göteborg, Sweden 2 Department of Molecular and Clinical Medicine, Sahlgrenska University Hospital and Academy, Göteborg University, Göteborg, Sweden

L. Andersson, Department of Occupational and Environmental Medicine, Sahlgrenska University Hospital and Academy, PO Box 414, S-405 30 Göteborg, Sweden. Tel: +46-31-786-28-47; Fax: +46-31-40-97-28; E-mail: lena.andersson{at}amm.gu.se



   Abstract
Top
 Abstract
Introduction
 Subjects and methods
 Results
 Discussion
 References
 
Background. Alpha-1-microglobulin (A1M) is a low molecular weight protein that can be measured in urine and used as a marker for tubular function, assuming that the normal variability within and between individuals is known. The aims of this study were to investigate this variability, to find the optimal way of sampling and quantifying A1M in spot urine samples to reflect the 24 h excretion and to examine storage stability.

Method. Timed urine specimens were collected from 29 healthy volunteers at fixed time points over 24 h on two separate days. Volumes, creatinine and specific gravity were determined. All samples were analysed with a commercial ELISA for A1M.

Results. We found a clear diurnal variation in A1M excretion rate and a gender effect (higher in males). The excretion rate was higher in the daytime, with high urinary flow, compared to overnight values. A1M excretion in spot urine samples was highly correlated with the 24 h excretion at all times except 22:00 in male subjects. Urinary A1M adjusted for creatinine concentration correlated well with the 24 h excretion. Variability within individuals was only 20% of the total variability in 24 h A1M excretion, but 43% in first morning urine. Expressed as CV, the intra-individual variability (between days) was 29% in 24 h excretion.

Conclusion. We conclude that diurnal variation and gender should be taken into account when comparing groups. Moreover, in spot samples (e.g. first morning samples) adjustment of A1M for creatinine or specific gravity is a reliable alternative to 24 h urine.

Keywords: Alpha-1-microglobulin; diurnal; protein HC; urine sampling; variability



   Introduction
Top
 Abstract
 Introduction
Subjects and methods
 Results
 Discussion
 References
 
Forty years ago, alpha-1-microglobulin (A1M) was discovered and named after its size and electrophoretic migration, as was the practice with other plasma proteins. It is also known as protein HC (human complex-forming glycoprotein, heterogeneous in charge). The physiological role of the lipocalin A1M is immunoregulatory in a suppressive way, and it is considered to be an anti-oxidant that can scavenge pro-oxidant haem groups. It is both size- and charge-heterogeneous. The gene for A1M also codes for bikunin, a glycoprotein with a number of functions; it inhibits proteinases involved in metastasis of tumour cells, it blocks calcium uptake in cells and it modulates cell growth. A1M is produced in the liver, but is also present in many organs and in plasma, both in the free form and bound to IgA. Normally, free A1M does not exceed concentrations of >60 mg/l in plasma. Low molecular weight proteins such as A1M (27 kDa) pass freely through the glomerular membranes [1–3]. As a rule, about 99% of the free A1M in the primary urine is reabsorbed by the megalin receptor in proximal tubular cells, where it is then catabolized. However, small amounts are excreted in urine. Proximal tubular function can be evaluated by measuring the amount of low molecular weight (LMW) proteins in urine, since a compromised reabsorption by the proximal tubules will result in increased excretion. Increased A1M in urine can therefore be an early sign of renal damage, primarily on the proximal tubular level [4].

Several studies have been performed using A1M as a biomarker for tubular dysfunction, mostly in patients with renal damage. The urinary concentration of A1M will be affected by the urinary flow (UF) rate, with lower concentrations in ‘diluted’ samples. This problem of diluted or concentrated samples can be adjusted for by relating the amount of A1M to the excretion of creatinine [5]. Little is known, however, about diurnal variation of excretion rates, other sources of variability and optimal techniques for assessing A1M in epidemiological studies.

The aims of the present study were as follows:

(1) To examine diurnal variation and the effects of gender and UF rate on A1M excretion.
(2) To investigate how A1M in various types of spot samples reflects the 24 h excretion.
(3) To assess variability between and within individuals.
(4) To test the storage stability of A1M in frozen (–20°C) samples.



   Subjects and methods
Top
 Abstract
 Introduction
 Subjects and methods
Results
 Discussion
 References
 
Subjects
We examined 29 healthy (no diabetes, hypertension or kidney disease), non-smoking subjects: 14 women (mean age 39 years, range 23–56 years) and 15 men (mean age 35 years, range 23–59 years). The subjects were recruited from the staff of our department and among students at Göteborg University. Each subject filled in a questionnaire about age, weight, smoking habits and any diseases or medication. The study was approved by the Ethics Committee of Göteborg University.

Urine sampling
Samples were collected on two separate days (mostly 4–6 days apart) at six fixed times over 24 h: 09:30, 12:00, 14:30, 17:30, 22:00 and first morning urine. If urination was necessary between fixed times, the next container was used. All urine samples were timed, and the total volumes recorded. When the containers were returned, replicate specimens (5 ml each) were taken and transferred to Minisorb tubes (NUNC, Denmark). Samples were stored cold (4°C) until analysis (within 3 days). Replicates used for stability control were immediately frozen (–20°C), and analysed after 2 weeks, 2 months and 6 months. These samples were thawed at room temperature and vortexed for 5 s before analysis [6].

Determination of A1M, creatinine and specific gravity
Analyses of A1M were performed using the {alpha}1-microglobulin ELISA Kit K6710 (Immundiagnostik AG, Bensheim, Germany). After washing of the 96-well plate, 100 µl NaCl and 10 µl of standard or sample solutions were incubated for 1 h in microtiter wells coated with rabbit polyclonal antibodies against human A1M. After washing, peroxidase-labelled antibodies against A1M were added. After 1 h of incubation and a final washing, the conjugate was reacted with H2O2-tetramethylbenzidine. The reaction was stopped by adding an acidic solution before absorbance was measured at 450 nm. The absorbance is proportional to the concentration of A1M. Calibrators with target values in the ranges 0.09–0.28 mg/l, provided in each kit, were used at the start and end of the assay in every run; our results were always within the accepted range. The coefficient of variation (CV), as calculated from duplicate analyses of the calibrators, was 11%. For duplicate analyses of samples performed on four occasions within 6 months, the median CV was 17 (13–38%). The limit of detection as calculated from Ablank + 3 xSDblank was 0.1 mg/l.

Analyses of creatinine were performed in fresh urine using the Jaffé method (Roche Diagnostics, limit of detection 0.01 mmol/l). Specific gravity was measured in fresh urine with a Ceti, Digit 012 refractometer (Medline, Oxfordshire, UK).

Data analysis
Excretion rates of A1M were calculated from concentration, volume and sampling time for each urine portion. All urine samples were corrected for creatinine and SG. Specific gravity calculations were performed with SGstandard = 1.017.

A1M levels were found to be skewed, mainly owing to two outliers (one man and one woman). Differences between genders and sampling times (means of 2 days) were tested with t-tests on ln-transformed levels. Associations between A1M excretion rates and UF rates were tested using Wilcoxon Signed Rank Test on Pearson's correlation coefficients. For some analyses, we used a mixed effects model (PROC MIXED, SAS Institute, Inc., Cary, NC, USA) to test the effects of time, flow rate, body weight and gender on A1M excretion. The total variability in A1M was partitioned within and between subjects using nested analysis of variance (PROC NESTED, SAS). The variability in A1M within individuals was also expressed as the coefficient of variation for excretion rates and creatinine-corrected concentrations on different days in each subject. The A1M excretion rates, as well as concentrations corrected for creatinine or SG at fixed times, were compared with the 24 h A1M excretion using linear regression and Pearson's correlation coefficient. For values below the limit of detection (LoD), the calculated value of the LoD divided by the square root of 2 was used in the statistical calculations [7].



   Results
Top
 Abstract
 Introduction
 Subjects and methods
 Results
Discussion
 References
 
Diurnal variation and effect of gender
First morning urine samples had the lowest excretion rates, significantly different from those in any of the daytime samples (P < 0.01). Men had significantly higher (about 80%) 24 h excretion rates than women (P = 0.012) (Table 1). The diurnal variation is illustrated in Figure 1, stratified by sex. Excretion rates differed significantly between men and women in first morning urine, at 09:30 and at 22:00.


View this table:
[in this window]
[in a new window]

  		Table 1 Excretion rates and urinary flow rates for each fixed time in 15 men and 14 women. Means and medians were calculated from the 2-day average excretion rates and urinary flows

 

Figure 1
View larger version (18K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Each bar represents the mean value and 95% confidence interval for the excretion of A1M at different times of the day in men and women. Excretion rate in first morning urine was significantly lower than at the other points of time for women, and the excretion rate at 09:30 was significantly different from that at other times of the day in men.

 
Effect of urinary flow
The effect of UF on the excretion of A1M was first assessed by examining the association between excretion rates and urinary flows (6 x 2 urine samples per individual) in each subject. The associations tended to be positive, with a median Pearson's correlation coefficient of 0.29 (N = 29; range –0.44 to 0.76)—significantly different from zero (P = 0.006). However, stratification for gender showed that the positive association existed only in men. Second, we examined the effect of UF, time, gender and body weight on A1M in mixed effects models using the 12 observations in each of the 29 subjects. The intercept was a random effect, and gender, time (dichotomized into first morning urine versus daytime samples), body weight and UF were fixed effects. Interaction terms (gender*UF and gender*time) were also tested. All these models showed a statistically significant effect of time (P < 0.0001). In models with either time and gender*UF or time and body weight, these additional factors were statistically significant, but the model with gender*UF had the best fit (measured as the Aikaike Information Criterion). However, data adjusted for creatinine concentrations showed no statistically significant effect of gender*UF.

Spot urine samples versus 24 h collections
The associations between the excretion rates of A1M at the six fixed times and the 24 h excretion were strong and similar over time (rp = 0.75–0.94) (Figure 2) except for male subjects at 22:00 (rp = 0.42). One male and one female subject had values that were more than three times higher than the medians for men and women, respectively. When these outliers were excluded, the correlation coefficients were still high but less strong (rp = 0.41–0.89).


Figure 2
View larger version (12K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Associations between excretion rates at 17:30 and 24 h excretion (a) and between first morning urine and 24 h excretion (b). A male outlier was excluded with A1M 0.47 mg/h at 17:30 and 0.52 mg/h in 24 h urine in (a) and with A1M 0.19 mg/h in first morning urine and 0.52 mg/h in 24 h urine in (b).

 
Variability within and between individuals
Of the total variability in 24 h A1M excretion, 80% was variability between individuals and 20% was variability within individuals. In first morning urine, 57% was variability between individuals (data for creatinine-corrected levels given below). Since there was a clear difference in excretion rates due to sex, variability between and within individuals was also calculated for each sex. Then, as could be expected, the interindividual variability constituted a somewhat smaller part of the total variability (e.g. for first morning urine: 54% for women and 32% for men). In general, the intraindividual variability was higher in daytime samples. Expressed as CV, the intraindividual variability was 29% in the 24 h A1M excretion, and 57% in first morning urine (86% for women and 39% for men).

Storage
Forty samples were used to assess storage stability. After 2 weeks, 2 months and 6 months, no significant decreases in concentrations were seen (point estimates 109, 124 and 93% of the initial concentrations), but the imprecision was relatively higher (CVs of 40, 49 and 40% respectively) in paired results from fresh and thawed samples.

Adjustment for creatinine or specific gravity
When A1M levels were adjusted for creatinine concentrations, the diurnal effect remained, but the effect of gender decreased somewhat (Figure 3). Creatinine-corrected levels of A1M at any time of the day were highly correlated to the 24 h excretion (r = 0.83–0.96) (Figure 4a). In addition, A1M corrected for specific gravity was associated with the 24 h excretion of A1M, albeit less strongly (r = 0.31–0.96) (Figure 4b). Table 2 presents coefficients of variation for three different ways of expressing A1M in 24 h urine. The CVs were similar despite different expressions; however, unadjusted concentrations slightly increased the variability (data not shown). For creatinine-corrected excretion in first morning urine, partitioning of the total variability showed that only 17% was variability between individuals (0% in women and 53% in men when stratified for gender).


Figure 3
View larger version (18K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Each bar represents the mean value and 95% confidence interval for the excretion of A1M in mg/g creatinine at different times of the day in men and women.

 

Figure 4
View larger version (13K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Associations between adjusted A1M excretion in first morning urine and 24 h excretion, where a represents excreted A1M adjusted for creatinine concentrations and b represents excreted A1M adjusted for specific gravity. A male outlier was excluded with A1M 2.52 mg/g creatinine and 0.52 mg/h in 24 h urine and with 2.64 for SG corrected A1M and 0.52 mg/h in 24 h urine.

 

View this table:
[in this window]
[in a new window]

  		Table 2 Intraindividual variability, expressed as the coefficient of variation (CV) in percent, for A1M in paired 24 h urine from two different days in 29 subjects

 


   Discussion
Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
References
 
The renal excretion of the low molecular weight protein alpha-1-microglobulin has been studied in patients with renal disease as well as in healthy subjects exposed to nephrotoxicants [8–14]. However, most studies have been performed on spot urine samples, and few studies discuss sampling optimization. In the present study, the excretion of A1M was assessed using repeated timed urine samples over the day and in 24 h urine in healthy men and women.

There was a pronounced diurnal variation with higher excretion rates in daytime samples. Factors that may have contributed to the diurnal variation include the higher UF during the day, body position, activity and changes in glomerular filtration rate. High UF could also decrease the fractional tubular reabsorption. The statistical analysis showed a tendency towards an increase in A1M excretion (mg/h) with UF rate, in addition to the diurnal variation, but only in men, and the effect of flow rate in men was not significant for creatinine-corrected results. However, previous studies have shown that the excretion rates of other proteins, as well as creatinine, are affected by UF rate [15,16]. It has been recommended that urine samples with very high or very low flow rates should be discarded, since they may not be representative. Separate calculations (data not shown) confirmed that when six subjects with flow rates <0.3 ml/min were excluded, the variability between individuals decreased moderately. It is well known that strenuous activity may induce proteinuria, especially microalbuminuria, but also tubular proteinuria [17]. The lack of information on each individual's activity meant that we could not investigate this any further. However, despite the diurnal variation, the excretion rates of A1M correlated well with the 24 h excretion at all times except at 22:00.

The difference in excretion rates between men and women has been reported previously [9] and could mainly be explained by body weight and larger kidneys in men. However, as shown in Figure 2, there was also a slight difference in creatinine-corrected levels. Our subjects were 21–59 years of age, and we cannot exclude that A1M excretion is different in adolescents or older subjects.

The excretion rates of A1M (mg/h) were highly correlated to levels adjusted for creatinine, and this was usually also the case for the 24 h excretion. We therefore conclude that the use of creatinine-adjusted spot samples is justifiable. Spot samples are easier to use than timed samples, and there are fewer problems with compliance. Adjustment for specific gravity (often used as an alternative to adjustment for creatinine) also seems to be reliable, and the coefficient of variation is close to that for creatinine adjustment.

The variability between individuals and between days within individuals is an important factor to consider in study design. Low variability between individuals is an advantage when comparing different groups, for example patients or groups exposed to nephrotoxic factors versus control groups. However, in studies where subjects serve as their own controls, for example in experimental studies, low variability within individuals is more important. Since the use of 24 h urine samples may be difficult in many settings, we recommend first morning urine samples, adjusted for creatinine, which have moderate variability both between and within individuals.

As has been reported by others [6], A1M is stable in frozen urine (–20°C) for at least 6 months without preservatives. However, the imprecision is higher in these samples, possibly due to the thawing process. It is important to vortex frozen samples to homogenize the sample. Adding preservatives, or storage at –70°C may increase storage stability even further [6].

We consider the sampling procedure reliable in the present study. The urinary volumes seemed adequate with no missing values. The analytical imprecision was acceptable for the calibrators, but especially on one occasion it was higher than expected in the subjects’ samples; this was partly due to one outlier and may have been caused by inhomogeneous urine samples. However, this source of variability is smaller compared to the true intra- and interindividual variability, and it will not affect our results. Our findings should be relevant to research performed on healthy people exposed to nephrotoxicants. The excretion of A1M has been shown to increase also in patients with clinical kidney disease and proteinuria [4,18]. We tested this in five female patients with glomerulonephritis and proteinuria (24 h albumin excretion 0.4–4.5 g), using the same assay as in our healthy individuals, and found increased A1M in 3/5 patients (7–18 mg/g creatinine). We did not, however, have access to repeated timed samples, and therefore we cannot know whether our findings on diurnal variation and other sources of variability are also valid for patients with clinical kidney disease and proteinuria.

In summary, despite the diurnal variation and gender differences reported above, determination of urinary A1M as a marker of tubular function poses few problems, and the 24 h excretion of A1M is reasonably well reflected by A1M in a first morning urine sample adjusted for either creatinine or specific gravity.



   Acknowledgments
 
Eva M. Andersson is acknowledged for statistical advice.

Conflict of interest statement. None declared.



   References
Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
 References
 

  1. Ekstrom B, Peterson PA, Berggard I. A urinary and plasma alpha1-glycoprotein of low molecular weight: isolation and some properties. Biochem Biophys Res Commun (1975) 65:1427–1433.[CrossRef][Web of Science][Medline]
  2. Allhorn M, Berggard T, Nordberg J, et al. Processing of the lipocalin alpha(1)-microglobulin by hemoglobin induces heme-binding and heme-degradation properties. Blood (2002) 99:1894–1901.[Abstract/Free Full Text]
  3. Penders J, Delanghe JR. Alpha 1-microglobulin: clinical laboratory aspects and applications. Clin Chim Acta (2004) 346:107–118.[CrossRef][Web of Science][Medline]
  4. Bakoush O, Grubb A, Rippe Tencer B J. Urine excretion of protein HC in proteinuric glomerular diseases correlates to urine IgG but not to albuminuria. Kidney Int (2001) 60:1904–1909.[CrossRef][Web of Science][Medline]
  5. Carrieri M, Trevisan A, Bartolucci GB. Adjustment to concentration-dilution of spot urine samples: correlation between specific gravity and creatinine. Int Arch Occup Environ Health (2001) 74:63–67.[CrossRef][Web of Science][Medline]
  6. Tencer J, Thysell H, Andersson K, et al. Long-term stability of albumin, protein HC, immunoglobulin G, kappa- and lambda-chain-immunoreactivity, orosomucoid and alpha 1-antitrypsin in urine stored at –20°C. Scand J Urol Nephrol (1997) 31:67–71.[Web of Science][Medline]
  7. Hornung RW, Reed LD. Estimation of average concentration in the presence of nondetectable values. Appl Occup Environ Hyg (1990) 5:46–51.
  8. Jarup L, Alfven T. Low level cadmium exposure, renal and bone effects—the OSCAR study. Biometals (2004) 17:505–509.[CrossRef][Web of Science][Medline]
  9. Kouri T, Harmoinen A, Laurila K, et al. Reference intervals for the markers of proteinuria with a standardised bed-rest collection of urine. Clin Chem Lab Med (2001) 39:418–425.[CrossRef][Web of Science][Medline]
  10. Hjorth L, Helin I, Grubb A. Age-related reference limits for urine levels of albumin, orosomucoid, immunoglobulin G and protein HC in children. Scand J Clin Lab Invest (2000) 60:65–73.[CrossRef][Web of Science][Medline]
  11. Mantur M, Kemona H, Dabrowska M, et al. alpha1-microglobulin as a marker of proximal tubular damage in urinary tract infection in children. Clin Nephrol (2000) 53:283–287.[Web of Science][Medline]
  12. Jung K, Pergande M, Graubaum HJ, et al. Urinary proteins and enzymes as early indicators of renal dysfunction in chronic exposure to cadmium. Clin Chem (1993) 39:757–765.[Abstract/Free Full Text]
  13. Pfleiderer S, Zimmerhackl LB, Kinne R, et al. Renal proximal and distal tubular function is attenuated in diabetes mellitus type 1 as determined by the renal excretion of alpha 1-microglobulin and Tamm–Horsfall protein. Clin Investig (1993) 71:972–977.[Web of Science][Medline]
  14. Weber MH, Verwiebe R. Alpha 1-microglobulin (protein HC): features of a promising indicator of proximal tubular dysfunction. Eur J Clin Chem Clin Biochem (1992) 30:683–691.[Web of Science][Medline]
  15. Vree TB, Hekster YA, Hafkenscheid JCM, et al. The influence of urine flow on renal clearance of creatinine in patients with normal and impaired kidney function. Drug Intell Clin Pharm (1981) 15:194–198.[Medline]
  16. Wetzels JF, Hoitsman AJ, Huvmans FT, et al. Influence of urinary flow-rate on proteinuria. Scand J Clin Lab Invest (1989) 49:195–197.[Web of Science][Medline]
  17. Poortmans JR, Mathieu N, De Plaen P. Influence of running different distances on renal glomerular and tubular impairment in humans. Eur J Appl Physiol Occup Physiol (1996) 72:522–527.[CrossRef][Web of Science][Medline]
  18. Holdt-Lehmann B, Lehmann A, Korten G, et al. Diagnostic value of urinary alanine aminopeptidase and N-acetyl-β-D-glucosaminidase in comparison to {alpha}1-microglobulin as a marker in evaluating tubular dysfunction in glomerulonephritis patients. Clinica Chimica Acta (2000) 297:93–102.[CrossRef][Web of Science][Medline]
Received for publication: 15. 5.07
Accepted in revised form: 18. 9.07


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
23/4/1252    most recent
gfm729v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Andersson, L.
Right arrow Articles by Barregard, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Andersson, L.
Right arrow Articles by Barregard, L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?