Evaluation of four commercially available assays for free thyroxin

We assessed two two-step and two analog assays for measuring free thyroxin (FT4) in serum: Clinical Assays' "GammaCoat Free/Total T4" (CA), Vitek's "KinetiCount Phase II Free T4" (VTK), Diagnostic Products Corporation's "Coat-a-Count Free T4" (DPC) kit (June 1987 version), and Amersham's "Amerlex-M Free T4" (AMX). The VTK assay is automated except for the initial pipetting step. Interassay results correlated well except for samples with abnormal serum albumin concentrations. FT4 values for hypoalbuminemic samples showed a highly significant (P less than 0.0001) correlation with serum albumin concentration in the DPC and AMX assays. The relationships are described by the equations y = 0.382albumin (g/L) + 0.81 pmol/L and y = 0.450albumin (g/L) - 3.20 pmol/L, respectively. When we used an equation derived from the Law of Mass Action to adjust FT4 values to values expected at an ideal albumin concentration, the observed correlation of albumin and FT4 was abolished completely in the DPC assay, and partly so in the AMX assay. The precision of CA was comparable with that of the analog assays; the CV for the VTK assay was approximately twice that for the other three assays.