Comparison of urinary albumin quantification by immunoturbidimetry, competitive immunoassay, and protein-cleavage liquid chromatography-tandem mass spectrometry.

BACKGROUND Increased urinary albumin excretion is a well-documented diagnostic and prognostic biomarker for renal disease. Urinary albumin is typically measured in clinical settings by immunoassay methods. However, neither a reference method nor a urine albumin calibration reference material is currently available. METHODS We quantified urinary albumin in patient samples by using 3 commercially available reagent systems: DiaSorin SPQ and Beckman Coulter LX 20 (immunoturbidimetric), and Siemens Immulite (competitive immunoassay). Results were compared to values obtained by protein-cleavage liquid chromatography-tandem mass spectrometry (LC-MS/MS). RESULTS In general, results from the 3 immunoassays agreed with results from LC-MS/MS. However, the SPQ results showed a negative bias across all ranges of albuminuria [(0-200 mg/L, y = 0.91x - 3.74 (CI 0.86-0.96); > 200 mg/L, y = 0.88x - 40.30 (CI 0.76-1.00)], whereas the LX 20 showed minimal bias in the 0-200 mg/L range [y = 0.97x - 88 (CI 0.92-1.02)] and the Immulite assay showed positive bias in the 0-200 mg/L range [y = 1.15x - 4.38 (CI 1.09-1.20)]. CONCLUSIONS These results showed a reasonable quantification of urinary albumin by representative polyclonal and monoclonal immunoassays compared to an LC-MS/MS assay. In addition, the results do not suggest the presence of nonimmunoreactive albumin in urine. However, differences in analytic performance between assays support the need for a reference calibration material and reference method to standardize clinical laboratory measurements of urinary albumin.