Performance evaluation of different albumin assays for the detection of analbuminemia.

ABSTRACT

Analbuminemia is characterized by the near absence of albumin in the plasma. Different methods are available for measuring albumin levels, but they do not necessarily agree with one another. It is a concern that analbuminemic samples could be falsely characterized due to the incorrect estimation of albumin. The objective of the work was to evaluate the performance of different assays in detecting analbuminemia. Albumin knockout (Alb-/-) mouse plasma was used to test the suitability of different albumin assays for their ability to properly characterize extreme albumin deficiency. Bromocresol green (BCG), bromocresol purple (BCP), enzyme-linked immunosorbent assay (ELISA), liquid chromatography-tandem mass spectrometry (LC-MS/MS), and gel electrophoresis were tested. The LC-MS/MS assay exhibited broad coverage of the amino acid sequence of albumin and indicated 8,400-fold lower (P<0.0001) albumin expression in Alb-/- than wildtype (WT), demonstrating its suitability for identifying extreme albumin deficiency. ELISA estimated albumin at 1.5±0.1 g/dL in WT and was below the detection limit in all Alb-/- samples. Gel electrophoresis yielded consistent results with LC-MS/MS and ELISA. The BCG assay overestimated albumin with apparently appreciable albumin concentrations in Alb-/- mice, yet the assay still indicated a significant difference between genotypes (Alb-/-, 1.2±0.05 g/dL, WT, 3.7±0.1 g/dL, P<0.0001). BCP drastically overestimated albumin and could not successfully identify the known analbuminemic phenotype of Alb-/- mice. By using Alb-/- plasma as a reference material and LC-MS/MS as a reference method, ELISA and gel electrophoresis appear appropriate for identifying analbuminemia, while BCG and BCP are not suitable. It is concluded that dye-binding assays should be avoided when extreme hypoalbuminemia or analbuminemia is suspected.