SubscribePantelis A. Sarafidis1, Jessica Riehle1, Emad Basta1, Atul Chugh1, and George L. Bakris2.
1) Hypertension/Clinical Research Center, Department of Preventive Medicine, Rush University Medical Center, Chicago, IL.
2) Hypertensive Diseases Unit, Section of Endocrinology, Diabetes & Metabolism, University of Chicago-Pritzker School of Medicine.
Background
Microalbuminuria is widely recognized as marker of abnormal vascular function evident at the product of the kidney, the urine [1], and has long been considered as the first indication of renal injury in patients with diabetes [2]. Accumulating evidence also suggests that microalbuminuria is associated with higher risk for cardiovascular events, as well as cardiovascular and overall mortality independently from traditional risk factors both in the general population and in patients with other risk factors or prevalent cardiovascular disease [3].
Thus, screening for microalbuminuria is currently recommended for all diabetic patients [2], but it can also provide valuable information for all individuals at high cardiovascular risk. The traditional way to detect abnormal urine albumin excretion (UAE) was to measure total urine albumin in 24-hour or other timed collections; measuring the albumin-to-creatinine ratio (ACR) in a spot collection of morning urine in the fasting state was recently recommended as a simple, quick and comparably accurate way of determining UAE [2;4].
Development of devices that can provide ACR estimations within some minutes by simultaneously measuring urine albumin and creatinine has probably helped expanding the use of ACR for microalbuminuria detection. However, the use of this approach requires knowledge of the factors that can affect spot ACR measurement [1].
Apart from the large intra-individual variation of UAE both in a single day and from day to day [5], factors related to urine creatinine can affect spot ACR measurement; the amount of muscle mass directly affects the level of creatinine in the urine and thus, caution is required when interpreting the ACR in patients with higher muscle mass, i.e. males or African Americans [6;7] or lower muscle mass, i.e. elderly people.
In addition, the analytical quality of the method used for creatinine determination will influence ACR estimations, and there is also the economical aspect of the equation, as ACR requires an additional creatinine analysis.
The HemoCue Albumin 201 system is a newly developed fast an easy way of quantitative screening for microalbuminuria at the point-of-care or the laboratory. It consists of a portable photometer and standard optical measuring cuvettes and measures urine albumin with an immunoturbidimetric method.
Figure 1. Correlation plot between urine albumin measured with the HemoCue Albumin 201 system and urine albumin measured with the reference method, y = 0.9978x - 1.0217, R2 = 0.904, N=155.
Objective
The aim of this study was to compare the validity of the HemoCue Albumin 201 system and the albumin-to-creatinine (ACR) method against a standard laboratory method of assessing urine albumin excretion (UAE).
Materials and Methods
Subjects
In this study we determined the levels of UAE in urine samples of 165 subjects with various risk factors for cardiovascular and chronic kidney disease. All subjects were attending the Hypertension Outpatient Clinic of the Hypertension/Clinical Research Center, Department of Preventive Medicine at Rush University Medical Center, Chicago, IL. The study was approved by the Rush University Institutional Review Board and all participants signed an informed consent prior to study initiation.
Study Protocol
All subjects were invited to participate in the study during their scheduled visits in the Outpatient Clinic of the Hypertension/Clinical Research Center, at Rush University Medical Center. After the purpose of the study was explained to subjects and they gave informed consent, a brief interview followed, where information on patient demographics and medical history was recorded. All participants then gave either a random urine specimen or a first-morning urine specimen for UAE screening.
Urine albumin was determined at the study site with the HemoCue Albumin 201 system (HemoCue, Ängelholm, Sweden). In addition, urine albumin and creatinine were measured at a central laboratory with standard laboratory methods. The study tested compared the validity of the HemoCue Albumin 201 system and the ACR method, as well as the sensitivity, specificity, positive predictive value [PV(+)] and the negative predictive value of these methods for the diagnosis of elevated UAE against the albumin measurement at the central laboratory. For elevated UAE we used a cut-off point of 20 mg/L for urine albumin estimated with the HemoCue Albumin 201 system cut-off points of 2.5 and 3.5 mg/mmol (for females and males respectively) for ACR.
Figure 2. Correlation plot between albumin-to-creatinine ratio and urine albumin measured with the reference method, y = 0.0815x + 0.3373, R2 = 0.784, N=155.
Laboratory Analyses
The HemoCue Albumin 201 system was designed for the quantitative determination of urine albumin levels at the point-of-care with an immunoturbidimetric reaction. It consists of a portable photometer and standard disposable optical measuring cuvettes. Each cuvette consists of a body with a cavity, which holds exactly 18 μL and contains reagent (a specific rabbit polyclonal anti-human albumin antibody) deposited on its inner walls. The urine sample is drawn into the cavity by capillary action.
The filled cuvette is inserted into the HemoCue Albumin 201 analyzer where the contents of the cuvette are mixed through vibration. Within 90 seconds, the immunochemical reaction is completed and the turbidity is measured photometrically at 610 nm.
The measuring range is of the device is 5-150 mg/L. Values below 5 are displayed as LLL, values above 150 mg/L are displayed as HHH. At the central laboratory, urine albumin was measured with an immunoturbidimetric method and creatinine with a kinetic modification of the Jaffe method on an Olympus analyzer (Olympus Life and Material Science Europa GmbH, Hamburg, Germany). Albumin-to-creatinine ratio (in mg/mmol) was calculated by dividing the level of urine albumin (in mg/L) with the level of urine creatinine (in mmol/L) in each sample.
Statistical analysis
Statistical analysis was performed using the using the SAS version 8.20 software (SAS Institute Inc, Cary, NC, USA). The validity of the HemoCue Albumin 201 system and the ACR was assessed by performing linear regression analyses and calculating the Pearson's coefficients of correlation (r) and coefficients of determination (r2) between each of these two methods and the reference method. We also build the respective Bland-Altmann plots (i.e. the difference between the corresponding measurements plotted against the average of the two measurements) [8].
The sample of this study had 90% power to detect correlations at the 0.25 level and much higher power for any correlations above that level. For reasons of calculation, all values below the low detection limit of the HemoCue system (5 mg/dL) are expressed as 3 mg/L. Values above the detection limit were excluded from the correlation analysis, but included in the analyses of diagnostic accuracy.
Figure 3. Bland-Altman plot of the HemoCue Albumin 201 system urine albumin measurement versus the reference method measurement.
Results
Baseline demographic data of the subjects whose urine samples were used in this analysis are presented in Table 1. All of the 165 urine samples obtained were tested with the Hemo-Cue Albumin 201 system at the study site and were analyzed at the central laboratory.
As depicted in Fig. 1, both the UAE levels measured with the HemoCue Albumin 201 system and the ACR were strongly correlated with the respective albumin values obtained with the reference laboratory method. Linear regression analysis between the results of the HemoCue Albumin 201 and the reference method produced the regression line y = 0.9978x - 1.0217, R2 = 0.904, and between ACR and urine albumin measured with the reference method produced the regression line y = 0.0815x + 0.3373, R2 = 0.784.
Thus, the UAE levels measured with the HemoCue system exhibited a slightly higher correlation with the results of the reference method (r=0.951, P<0.001) compared to the ACR measurements (r=0.885, P<0.001). As shown in the Bland-Altman plot in Fig. 2, there was good agreement between the HemoCue UAE measurements and the measurements of the reference method. In the population of the present study the prevalence of elevated UAE (determined with the reference method) was 24%.
The HemoCue Albumin 201 system showed a sensitivity of 92% and a specificity of 98% for the diagnosis of elevated UAE when compared to the reference method (Table 1). Estimation of ACR displayed sensitivity of 73% and specificity of 96% for the diagnosis of elevated UAE. The PV(+) and PV(-) of each test for the diagnosis of elevated UAE were 92% and 98% for the HemoCue system and 85% and 92% for ACR (Table 2).
Conclusions
The present study is the first to determine the validity and the diagnostic sensitivity, specificity, positive and negative predictive value of the HemoCue Albumin 201 system in comparison to ACR estimations.
Our findings strongly suggest that the HemoCue Albumin 201 system is at least as accurate as laboratory ACR estimations and has a slightly better combination of diagnostic sensitivity and specificity, positive and negative predictive value for microalbuminuria detection than ACR determination.
These results together with the simplicity of handling, the speed and the low cost of the HemoCue Albumin 201 device support the added value of this system for microalbuminuria detection at the point-of-care or the laboratory.