In women, no significant variances of calculated free testosterone (cFT) are observed when a fixed value of albumin (Alb: 4.3 g/dL) is used instead of measured albumin values.

Background: Calculated free testosterone (cFT) is determined from the values of total testosterone (TT), sex hormone binding globulin (SHBG), and albumin (Alb) using mathematical formulae. We evaluated any potential cFT variance when determined with fixed Alb (4.3 g/dL) compared to measured Alb, and the point at which low SHBG and Alb combinations produced significant cFT variance.

Materials: We analyzed 2050 data points in 1222 women. cFT values with fixed vs. the actual measured Alb values were evaluated and contrasted. cFT levels were determined theoretically for all possible combinations of TT, SHBG, and Alb.

Methods: Agreement between the two measures was assessed with Lin's concordance coefficient. Statistical analyses were performed using R software version 2.12.1.

Results: Mean Alb was 4.05 ± 0.30 g/dL. Mean SHBG 73.0 ± 53.3 nmol/L. A fixed Alb of 4.3 g/dL produced no significant variance for most evaluations of cFT. The accuracy decreased with Alb ≤ 3.5 g/dL in combination with SHBG ≤ 30 nmol/L and exists in 1.0% of the samples.

Conclusions: A fixed Alb of 4.3 g/dL is acceptable for most clinical evaluations. If Alb is ≤ 3.5 g/dL, along with SHBG ≤ 30 nmol/L, the variance increases and a free testosterone (FT) measurement by equilibrium dialysis is warranted for better accuracy.