Laboratory issues in measuring and reporting urine albumin
1 Department of Pathology, Virginia Commonwealth University, Richmond, VA 23298-0286, USA 2 Department of Pathology, University of Virginia Medical School, Charlottesville, VA, 22908, USA
Correspondence and offprint requests to: W. Greg Miller, Department of Pathology, Virginia Commonwealth University, PO Box 980286; Richmond, VA, 23298-0286, USA. Tel: +1-804-828-0375; Fax: +1-804-828-0353; E-mail: gmiller{at}vcu.edu
Keywords: albumin:creatinine ratio; urine albumin
Urine albumin is an important biomarker for kidney damage, and its measurement is recommended by clinical practice guidelines in many countries for identifying and managing patients with kidney disease. A recent publication reviewed current practices in measurement and reporting of urine albumin concentrations and made recommendations for improvement [1]. The paper reflected the work of the Laboratory Working Group of the National Kidney Disease Education Program (NKDEP, USA) and the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) Joint Committee for Standardization of Albumin in Urine.
Historically, the concentration of albumin excreted in 24 h was used to evaluate kidney function. The difficulty of collecting 24 h urine samples has lead to the common recommendation to use an untimed urine collection and to report the ratio of the albumin concentration to the creatinine concentration. This albumin:creatinine ratio (ACR) is widely viewed as an acceptable surrogate for the albumin excretion rate. The recommendations from various professional organizations, however, vary regarding the type of urine samples to collect for measurement of ACR (e.g. first-morning void, random), the reporting units to use and the criteria for interpretation of the results. Most recommendations use a single decision threshold to identify patients with clinically significant albuminuria, despite the variation of both the albumin and creatinine excretion rates with time of day, exercise, gender, race and other factors. Further complicating the use of a single decision threshold is the increasing evidence that albumin excretion functions as a continuous variable in predicting a progressive increase in risk for complications related to decreased kidney function [2,3].
A number of urine collection issues need careful investigation before improved recommendations for practice can be developed. A wide range of estimates have been reported for the biologic variability of albumin and creatinine in urine samples collected at different times of the day. The evidence supports the conclusion that the ACR is less variable in a first morning sample than in a random sample. There is no evidence to determine if ACR in a second morning sample has advantages. Albumin is reported to adsorb to plastic surfaces, but no systematic evaluations have been published of the potential impact of adsorption on measured results. The stabilities of albumin and creatinine in urine after collection are not well understood. The influence of proteases, other chemical reactions and binding to sediment may compromise the sample during storage in the liquid state. The allowable sample storage time is unknown. Freezing at –20 °C is known to be unsatisfactory. Storage below –70 °C is recommended when the measurement cannot be performed promptly, but this is impractical in clinical practice.
The albumin molecule is not a clearly defined single entity in plasma or urine. Plasma albumin binds many ligands including fatty acids, bilirubin, hormones, ions such as calcium and chloride and many drugs. Albumin has a free thiol group and forms dimers with itself and conjugates with other molecules. Albumin has a half-life of 
20 days in circulation and is glycated to varying degrees depending on a patient's average glucose. Albumin fragments and truncated forms have been identified in plasma and in urine. Glomerular filtration is a size- and charge-dependent process, and various modified albumin forms may be either more easily or less easily filtered at the glomerulus. Receptor-mediated tubular reabsorption of albumin is likely influenced by the various molecular forms of albumin. The molecular forms of albumin in urine and serum are likely different after the action of proteases in the urinary tract. Moreover, urine has high concentrations of a number of ligands that bind to albumin.
All routine laboratory methods for measuring urine albumin are immunoassays. The range of potential molecular forms presents a variety of epitopes that may be different among individuals and among disease conditions. Albumin has at least five different antigenic sites, and some methods that use polyclonal antibodies have been shown to react with several degraded and fragmented species of albumin. The influence of albumin molecular forms on the measurement uniformity of contemporary routine methods has not been systematically evaluated. There have been reports that a ‘non-immunogenic’ form of albumin may be present in urine of diabetic patients and can be measured by use of high performance liquid chromatography (HPLC). Recent evidence, however, supports the view that the HPLC method measured non-albumin proteins that co-eluted with albumin [4,5]. Other evidence indicates that the HPLC test offers no advantages over immunoassay as a predictive marker for adverse cardiovascular events [6] or for renal or non-renal mortality [7].
Many countries use proficiency testing (PT), also called external quality assessment (EQA), in which a set of common samples is measured by a group of laboratories to evaluate the agreement of results among them. The primary limitation is that few programs provide urine samples that closely mimic a native patient's sample. The samples are supplemented with purified albumin, other analytes and stabilizers that may influence a method's results. Review of PT results from several countries showed that results from different laboratories using different methods had a CV of 15% at urine albumin concentrations above 20 mg/L. PT results from different laboratories that all used the same method had CVs of 5–10%. This agreement among methods may be adequate at higher concentrations, but will need improvement to enable development of risk criteria that extend to lower concentrations of urine albumin and that are stratified by age, gender and race.
The accuracy of routine methods for urine albumin and creatinine is unknown because there are no reference methods for albumin in serum or urine, and no reference materials for albumin or creatinine in urine. All manufacturers of routine urine albumin methods currently trace calibration to a serum albumin reference material (CRM 470, Institute for Reference Materials and Measurements, Belgium) that is diluted to the concentrations measured in urine. There is no standard procedure for dilution, and no standard diluent has been developed. Similarly, calibration of urine creatinine measurements frequently uses the same calibration as that for serum or plasma despite differences between urine and serum. A candidate reference material for urine albumin has been investigated by a Japanese group and a candidate reference measurement procedure based on isotope dilution mass spectrometry is in development by a group at Mayo Clinic (USA). Further improvement in accuracy of measurements will depend on these reference system components being credentialed by the Joint Committee for Traceability in Laboratory Medicine [8].
Current reporting practices for urine albumin and ACR can be improved. The terms ‘microalbumin’ and ‘macroalbumin’ should be discontinued and replaced with ‘urine albumin.’ The terms cause confusion, and practitioners often wonder if there is a small or large albumin molecule being excreted. The current terms were derived from the detection limits (‘sensitivity’) of certain urine dipstick tests, and have caused widespread use of arbitrary decision points that do not reflect the continuously increasing risk associated with progression of urine albumin concentrations. Furthermore, using these terms will prevent adoption of age-, gender- and race-specific interpretive guidelines for ACR. Differences in the units of reporting the ACR (such as mg/g and mg/mmol) cause further confusion and prevent worldwide adoption of uniform clinical practice guidelines. Units will be difficult to harmonize in all regions, but reporting units should be uniformly adopted within a country or geographic region. Laboratories should recognize that urine albumin concentration (e.g. mg/L) cannot be interpreted without information about the rate of excretion. Thus, the ACR should always be reported, with or without the albumin concentration.
The preceding summary of key issues includes several recommendations that can be put into practice to improve the uniformity of urine albumin collection and reporting practices. However, additional scientific evidence is needed in a number of areas to enable more complete recommendations to be developed for standardizing urine albumin and creatinine measurement and reporting. The NKDEP/IFCC Committee for Standardization of Albumin in Urine is working to develop the needed information, and to stimulate development of the critical reference system components to enable improved standardization of routine measurement systems.
The Laboratory Professionals section of the NKDEP web site [9] provides regular updates on the status of laboratory issues related to kidney disease.
Conflict of interest statement. The full report that is summarized in the editorial comment has been published by Clinical Chemistry.
 |
References |
|---|
 Top References |
|---|
- Miller WG, Bruns DE, Hortin GL, et al. Current issues in measurement and reporting of urinary albumin excretion. Clin Chem (2009) 55:24–38.
[Abstract/Free Full Text] - Linksde Jong PE, Curhan GC. Screening, monitoring, and treatment of albuminuria: public health perspectives. J Am Soc Nephrol (2006) 17:2120–2126.
[Abstract/Free Full Text] - Forman JP, Fisher NDL, Schopick EL, et al. Higher levels of albuminuria within the normal range predict incident hypertension. J Am Soc Nephrol (2008) 19:1983–1988.
[Abstract/Free Full Text] - Sviridov D, Meilinger B, Drake SK, et al. Co-elution of other proteins with albumin during size-exclusion HPLC: implications for urine albumin analysis. Clin Chem (2006) 52:389–397.
[Abstract/Free Full Text] - Shaikh A, Seegmiller J, Borland T, et al. Comparison between immunoturbidimetry, size-exclusion chromatography, and LC-MS to quantify urinary albumin. Clin Chem (2008) 54:1504–1510.
[Abstract/Free Full Text] - McQueen MJ, Gerstein HC, Pogue J, et al. Reevaluation by high-performance liquid chromatography: clinical significance of microalbuminuria in individuals at high risk of cardiovascular disease in the Heart Outcomes Prevention Evaluation (HOPE) study. Am J Kidney Dis (2006) 48:889–896.[CrossRef][Medline]
- Wang Z, Hoy WE, Nicol JL, et al. Predictive value of nephelometric and high-performance liquid chromatography assays of urine albumin for mortality in a high-risk aboriginal population. Am J Kidney Dis (2008) 52:672–682.[CrossRef][Web of Science][Medline]
- Database of higher-order reference materials, measurement methods/procedures, and services. http://www.bipm.org/jctlm/ (27 January 2009, date last accessed).
- National Kidney Disease Education Program. http://www.nkdep.nih.gov/labprofessionals (27 January 2009, date last accessed).
Accepted in revised form: 9. 1.09
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||