Study of a mechanism responsible for anomalous results for free thyroxin by some analog assays for subjects with albumin deficiency.

I examined the effect of albumin deficiency, as might occur in pregnancy and nonthyroidal illness, on four analog-based assays for free thyroxin (Amersham's Amerlex, Diagnostic Products Corporation's Coat-A-Count "one-step," Clinical Assays' Gammacoat "one-step," and Corning's Immophase "single-step"). In subjects with albumin deficiency per se, Amerlex and Coat-A-Count results for free thyroxin were not significantly different from normal, whereas free thyroxin values by Gammacoat and Immophase assays were lower than that of the control. Furthermore, in the albumin-deficient group, the mean amount of thyroxin associated with antisera in each assay was significantly higher than that of the control and correlated well with the degree of decrease in free thyroxin. I conclude that subnormal results for thyroxin by analog assays, in euthyroid patients with albumin deficiency, is a function of the specific assays and is not a characteristic of analog assays per se. In addition, the study suggests that albumin deficiency alters the distribution between thyroxin and its binders in all four analog assays, but only in the Gammacoat and Immophase methods is this alteration sufficiently large as to distort assay results.