A Critical Evaluation of Simple Methods for the Estimation of Free Testosterone in Serum

The free and nonspecifically bound plasma hormone levels generally reflect the clinical situation more accurately than total plasma hormone levels. Hence, it is important to have reliable indexes of these fractions. The apparent free testosterone (T) concentration obtained by equilibrium dialysis (AFTC) as well as the fraction of serum T not precipitated by 50% ammonium sulfate concentration (non-SHBG-T; SHBG, sex hormone-binding globulin), often referred to as bioavailable T, appear to represent reliable indexes of biologically readily available T, but are not well suited for clinical routine, being too time consuming. Several other parameters have been used without complete validation, however: direct immunoassay of free T with a labeled T analog (aFT), calculation of free T (FT) from total T and immunoassayed SHBG concentrations (iSHBG), and the free androgen index (FAI = the ratio 100T/iSHBG). In the view of substantial discrepancies in the literature concerning the free or bioavailable T levels, we compared AFTC, FT, aFT, FAI, and non-SHBG-T levels in a large number of sera with SHBG capacities varying from low, as in hirsute women, to extremely high as in hyperthyroidism. All these indexes of bioavailable T correlated significantly with the AFTC concentration; AFTC and FT values were almost identical under all conditions studied, except during pregnancy. Values for aFT, however, were only a fraction of either AFTC or FT, the fraction varying as a function of SHBG levels. Also, the FAI/AFTC ratio varied as a function of the SHBG levels, and hence, neither aFT nor FAI is a reliable index of bioavailable T. The FT value, obtained by calculation from T and SHBG as determined by immunoassay, appears to be a rapid, simple, and reliable index ofbioavailable T, comparable to AFTC and suitable for clinical routine, except in pregnancy. During pregnancy, estradiol occupies a substantial part of SHBG-binding sites, so that SHBG as determined by immunoassay overestimates the actual binding capacity, which in pregnancy sera results in calculated FT values that are lower than AFTC. The nonspecifically bound T, calculated from FT, correlated highly significantly with and was almost identical to the values of non-SHBG-T obtained by ammonium sulfate precipitation, testifying to the clinical value of FT calculated from iSHBG.