Free thyroxine assessed with three assays in sera of patients with nonthyroidal illness and of subjects with abnormal concentrations of thyroxine-binding proteins.

Three methods for estimating free thyroxine (FT4) in serum were studied: equilibrium dialysis, the SPAC-ET FT4 radioimmunoassay kit, and the Amerlite MAB FT4 luminometric assay. Serum samples from 10 subjects with above-normal thyroxine-binding globulin (TBG), 6 with low TBG, 30 with familial dysalbuminemic hyperthyroxinemia (FDH), 13 with nonesterified fatty acids (NEFA) concentrations in serum > 1.0 mmol/L, and 178 patients with various degrees of nonthyroidal illness (NTI) were measured and compared with samples from 42 euthyroid blood donors. The Amerlite MAB FT4 assay compared well with equilibrium dialysis, whereas the SPAC-ET assay averaged 40% lower. All three assays were not influenced by changes in TBG and showed no or only little changes in the presence of NEFA. Mean FT4 values in the FDH samples were somewhat higher than in controls when measured with the SPAC-ET assay, about equal with equilibrium dialysis, and somewhat below the mean control value with the Amerlite MAB FT4 assay, although individual results were within the control reference range. In NTI patients, no FT4 values were below the control reference range by the Amerlite MAB FT4 assay, 4 of 178 were below this range by equilibrium dialysis, and 1 of 178 was below this range by the SPAC-ET assay. In all assays a large proportion of the NTI samples showed FT4 values above the control reference range, a result that will interfere with the efficacy of these assays for assessing thyroid function in NTI patients.