Validation of albumin determined in urine with the HemoCue point‐of‐care analyser
H. von Schenck
Abstract
H. von Schenck
Abstract
Albumin concentrations persistently between 20 and 200 mg/L in first morning spot urine or urine collected overnight, referred to as microalbuminuria, indicate incipient nephropathy in diabetes mellitus. This study validates a new point-of-care device, the HemoCue Urine Albumin analyser, for handling, accuracy precision and predictive values (PV+/-) at 20 mg/L. Over a period of 2 months, 200 microalbuminuria samples were selected at the Department of Clinical Chemistry according to the results from the Integra 700 instrument (Roche, R) and analysed on the same day using the HemoCue analyser (HemoCue, H) and the Immage instrument (Beckman, B), in all cases closely following the manufacturers' instructions. Only 137 results were within the measuring range of H, 10-150 mg/L. Comparisons yielded regression lines H=1.06R-7.2 (r2=0.94), H=1.08B-3.1 (r2=0.94) and R=1.00B+4.3 (r2=0.99). Inter-assay (n=24) CV% at 12 mg/L was H=18.2, R=4.2 and B=2.9 and at 65 mg/L: H=6.1, R=1.8 and B=2.6. Intra-assay duplicate CV% for H at 21-40 mg/L was 13.2, at 41-80 mg/L 10.8 and at 81-150 mg/L 9.2. Intra-assay repeatability (n=8) CV% at 28 mg/L was 7.2-13.8, at 57 mg/L 6.4-8.4 and at 105 mg/L 4.3-7.1. External quality assurance urine albumin (B) was +5.7% cf. nephelometry and (R) +1.0% cf. turbidimetry (n=6) method-group means. PV+/- values were (H versus R) 0.98/0.37 and (H versus B) 0.95/0.65. HemoCue is easy to handle. Results below 20 mg/L need to be confirmed at the central laboratory to exclude microalbuminuria. Values above 20 mg/L can be used to follow microalbuminuria, as precision allows discerning steps of 10 mg/L.
OpenAlex reports 10 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Albumin concentrations persistently between 20 and 200 mg/L in first morning spot urine or urine collected overnight, referred to as microalbuminuria, indicate incipient nephropathy in diabetes mellitus. This study validates a new point-of-care device, the HemoCue Urine Albumin analyser, for handling, accuracy precision and predictive values (PV+/-) at 20 mg/L. Over a period of 2 months, 200 microalbuminuria samples were selected at the Department of Clinical Chemistry according to the results from the Integra 700 instrument (Roche, R) and analysed on the same day using the HemoCue analyser (HemoCue, H) and the Immage instrument (Beckman, B), in all cases closely following the manufacturers' instructions. Only 137 results were within the measuring range of H, 10-150 mg/L. Comparisons yielded regression lines H=1.06R-7.2 (r2=0.94), H=1.08B-3.1 (r2=0.94) and R=1.00B+4.3 (r2=0.99). Inter-assay (n=24) CV% at 12 mg/L was H=18.2, R=4.2 and B=2.9 and at 65 mg/L: H=6.1, R=1.8 and B=2.6. Intra-assay duplicate CV% for H at 21-40 mg/L was 13.2, at 41-80 mg/L 10.8 and at 81-150 mg/L 9.2. Intra-assay repeatability (n=8) CV% at 28 mg/L was 7.2-13.8, at 57 mg/L 6.4-8.4 and at 105 mg/L 4.3-7.1. External quality assurance urine albumin (B) was +5.7% cf. nephelometry and (R) +1.0% cf. turbidimetry (n=6) method-group means. PV+/- values were (H versus R) 0.98/0.37 and (H versus B) 0.95/0.65. HemoCue is easy to handle. Results below 20 mg/L need to be confirmed at the central laboratory to exclude microalbuminuria. Values above 20 mg/L can be used to follow microalbuminuria, as precision allows discerning steps of 10 mg/L.
Key concepts: Analyser, Urine, Point-of-care testing, Point of care, Albumin, Medicine, Chromatography, Chemistry