Abstract: Background-Aim: In clinical laboratories the determination of free and bioavailable testosterone is a challenge. The reference technique, equilibrium dialysis coupled to mass spectrometry, is not easy to apply in routine clinical practice. Therefore, the Vermeulen formula for free testosterone and the Morris formula for bioavailable testosterone, both based on immunoassays, are currently used and have shown good correlation with this method. The aim of our study is to assess whether there are differences in the results obtained from these formulae when different immunoassays are used. Methods: Albumin, testosterone and SHBG were analysed in 125 serum samples using the Siemens Atellica and Snibe Maglumi 8000 autoanalysers. In addition, free testosterone was determined by immunoassay on the Snibe analyser. Statistical analysis was performed with MedCalc software. Results: The results of bioavailable testosterone calculated by the Morris formula showed a mean of the differences in the Bland Altman plot of 9.40(2.75 - 16.05), in the Passing Bablok regression an ordinate at the origin of 0.40(-1.32 - 1.60) with a slope of 1.23(1.14-1.37) and a correlation coefficient of 0.927(0.898 - 0.948). Free testosterone results calculated by the Vermeulen formula showed a mean difference in the Bland Altman plot of 0.67 (0.001-1.34), in the Passing Bablok regression an ordinate at the origin of 1.25(1.09?1.46) with a slope of 2.30(2.14?2.56) and a correlation coefficient of 0.907(0.875?0.931). Comparison of free testosterone by immunoassay with the calculated results showed a mean on the Bland Altman plot of 7.40(5.60?9.20) when compared with that calculated using Maglumi8000 and 8.24(5.99?10.49) when compared with the calculation using Atellic. Conclusions: Our results show that both bioavailable and free testosterone are not interchangeable when different immunoassays are used for calculation. The laboratory should be aware of the limitations of immunoassays and consider the impact that a change in methodology may have on the results obtained from these formulae. Free testosterone measured by Snibe immunoassay is not equivalent with the formulae, even when the same analyser is used for the calculation.

Autoría

Descargar referencia