[Direct and indirect parameters for free thyroxine. II. Diagnostic accuracy in disorders of thyroid function and binding protein anomalies].

Part II of the study concerning the clinical applicability of direct and indirect parameters for free thyroxine evaluates the diagnostic accuracy of the FT4-RIAs ImmoPhase, GammaCoat, Liquisol, Amerlex and LisoPhase in relation to FT4 index and T4/TBG ratio. This comparison of methods is done on a thoroughly classified collection (n = 640) of patients with normal and impaired thyroid function including patients with binding protein anomalies (pregnancy, estrogen-medication, phenytoin therapy, renal protein loss). FT4 normal ranges of a given kit harmonize well with data of the manufacturers and of the literature. On the other hand, the normal ranges of the various kits are not comparable. The differentiation of euthyroidism from hyper- or hypothyroidism can be made without problems using any of these methods, with the exception of the FT4-RIA GammaCoat. As expected, patients with euthyroid goiter show, on the average, slightly lower parameters for FT4. In pregnancy all direct and indirect parameters for FT4 have a tendency to lower values after the first trimester. This trend is most distinct for the T4/TBG ratio. A weak negative correlation of FT4 parameters with basal TSH, which does not exceed the upper normal range, however, can be interpreted in the sense of a relative hypothyroxinemia. Under contraceptive estrogen medication FT4 parameters do not fall outside the normal range, with the exception of the FT4-RIA ImmoPhase assay which yields a significantly increased frequency of high FT4 levels. On therapy with phenytoin FT4 values are generally lower than in controls. FT4 parameters in patients with renal protein loss of more than 2 g daily do not behave uniformly. While three of the FT4-RIAs tested (ImmoPhase, GammaCoat, Amerlex) show a tendency to lower values, the results of the remaining FT4 parameters do not differ significantly from the distribution of normals. The study leads to the conclusion that indirect FT4 parameters are still useful in the diagnosis of thyroid function; currently available FT4 radioimmunoassays may yield comparable results. With regard to the varying quality of the various commercial test kits, the choice between FT4-RIAs and indirect FT4 parameters does not depend primarily on clinical but on technical viewpoints.