Is capillary zone electrophoresis a suitable method for estimating serum albumin: A comparison of four methods

The capillary zone electrophoresis method of albumin measurement is frequently used for oncologic and haematologic patients but few data exist about the agreement between the albumin measurements performed by capillary zone electrophoresis and other methods. The aim of this study was to analyse the agreement between human serum albumin measurements by capillary zone electrophoresis and by the nephelometry, bromocresol purple and turbidimetry methods.

We prospectively measured 100 freshly collected non-frozen patient serum samples, by using four different methods: the capillary zone electrophoresis method performed with a CAPILLARYS 2 instrument, the bromocresol purple dye method performed with an Advia XPT analyser, the nephelometric method performed with a BN ProSpec analyser and the turbidimetric method with reagents from DiAgam and performed with the Advia XPT analyser.

A bias towards higher values in the lower range of albumin concentrations was observed with capillary zone electrophoresis compared to immunonephelometry: correlation coefficient r² = 0.925; slope of 0.86 (0.82-0.89, 95% confidence interval), which is significantly different from 1; and an intercept of 4.94 g/L (3.67-6.16, 95% confidence interval). Similar results were observed when comparing capillary zone electrophoresis to the bromocresol purple and immunoturbidimetry methods. The capillary electrophoresis method overestimated low albumin levels by up to 25% (5 g/L).

Compared to the nephelometry, turbidimetry and bromocresol purple methods, the capillary zone electrophoresis method tends to overestimate human serum albumin concentrations for levels below 30 g/L. This discrepancy could lead to an overestimation of the nutritional status, an inappropriate scoring of the disease and a delay in nutritional treatment.