Three methods compared for estimating the fraction of testosterone and estradiol not bound to sex-hormone-binding globulin.

Here we propose and validate, both theoretically and experimentally, a new technique for measuring the percentage of non-sex-hormone-binding globulin (SHBG) bound to testosterone and estradiol. In this method, SHBG is saturated with 1 mumol of 5 alpha-dihydrotestosterone per liter and the non-protein-bound fraction is subsequently measured in treated and untreated samples by centrifugal ultrafiltration dialysis (CUFD). We compared this technique with two previously published methods for measuring the percentage of albumin-bound steroid--ammonium sulfate precipitation and heat denaturation of SHBG followed by CUFD--and also with previously published computer predictions. The ammonium sulfate precipitation gave results closest to the computer-predicted percentage of albumin-bound testosterone and estradiol. Of the two techniques involving CUFD, the 5 alpha-dihydrotestosterone saturation method was the more reproducible. Results by both CUFD methods were higher than the computer-based predictions for albumin-bound steroid in pregnancy serum, in part perhaps because the CUFD measurements represent the non-SHBG-bound fraction rather than that which is albumin-bound alone.