Evaluation of an algorithm for calculation of serum "Bioavailable" Testosterone (BAT)

Objectives: To evaluate a recently published algorithm for calculation of serum "Bioavailable" Testosterone (BAT) using serum Total Testosterone (TT), Sex Steroid Binding Globulin (SSBG) [also commonly known as Sex Hormone Binding Globulin (SHBG)] and albumin concentrations as parameters, in comparison with a locally available "salting-out" BAT method. If satisfactory, this calculation could serve as a substitute for the BAT assay, which would amount to a major cost saving and faster turnaround time. Design and Methods: During a 6-month period, 426 serum samples referred for BAT analysis to the Hospitals In-Common Laboratory of Toronto were also analyzed in-house for TT, SSBG and albumin for computation of comparison calculated BAT results. Results: A good statistical correlation was obtained, but only after unexpectedly drastic empirical modification of the association constant values: r = 0.95, Calculated %BAT = 0.971 x Measured %BAT + 0.008. The endocrinologist/andrologist of our team (JB), who was the responsible physician for all patients included in this study, reviewed the tabulated and charted calculated BAT results and verified that they were clinically equivalent. Conclusions: Although it is feasible to calculate BAT, the algorithm is not directly portable. Before adopting such a calculation each laboratory should compare it with the locally available BAT method and consider adjusting the calculation to optimize the correlation. Future reassessment may be necessary whenever the SSBG, TT or BAT assay is changed. (C) 2003 The Canadian Society of Clinical Chemists. All rights reserved.