Problems with the estimation of urine protein by automated assays.

OBJECTIVES Most clinical laboratories replaced their manual precipitation techniques for the determination of urinary protein with automated dye binding assays or benzethonium chloride-turbidimetric assays. Few studies have validated these assays for the measurement of urinary proteins in the normal range. DESIGN AND METHODS This study compares four automated assays for the measurement of urinary protein to a manual Ponceau S/TCA precipitation assay. We evaluated the linearity, the precision, the analytical sensitivity, the accuracy and the recovery of different proteins for each assay. RESULTS All assays showed good linearity with the theoretical concentration of albumin present in the sample. The coefficient of variation was below 10% at a concentration of 0.142 g/L. However, the manual Ponceau S/TCA assay demonstrated superior analytical sensitivity. Accuracy determinations showed a variable positive bias and poor correlations at concentrations below 0.1 g/L when compared to the Ponceau S/TCA assay. Small molecular weight peptides particularly affected the pyrogallol red assays but other urinary components also interfered with the automated assays. CONCLUSIONS Most automated assays show high imprecision and poor accuracy for the measurement of urinary protein in the normal range. The Ponceau S/TCA offers a precise and accurate manual alternative to these automated assays.