Is the Vermeulen formula most accurate for calculating free testosterone levels when Sex Hormone-Binding Globulin (SHBG) concentrations are between 15 and 70 nmol/L?

Accuracy of Vermeulen Formula for Free Testosterone Calculation Based on SHBG Levels

The Vermeulen equation for calculating free testosterone is most accurate when SHBG concentrations are above 30 nmol/L, with accuracy decreasing significantly when SHBG falls below this threshold. 1, 2

Accuracy of Calculated Free Testosterone Methods

• The accuracy of calculated free testosterone (cFT) using the Vermeulen equation is highly dependent on SHBG concentration, with decreased reliability when SHBG is below 30 nmol/L 1
• When using the Vermeulen equation, cFT calculations show high diagnostic accuracy with a pooled sensitivity of 0.89 and specificity of 0.83 when SHBG levels are within normal ranges 1
• The Free Androgen Index (FAI) also becomes inaccurate when SHBG concentrations fall below 30 nmol/L 1, 3
• When SHBG is ≤30 nmol/L and albumin is ≤3.5 g/dL (occurring in approximately 1.2% of cases), the variance in calculated free testosterone increases significantly 2

Gold Standard Measurement Methods

• Direct measurement of free testosterone using equilibrium dialysis is considered the gold standard and provides more accurate results than calculated methods, especially in cases with abnormal SHBG levels 1, 4
• Equilibrium dialysis should be used when SHBG is ≤30 nmol/L and albumin is ≤3.5 g/dL for better accuracy 2
• Functional measurement of free testosterone using equilibrium dialysis or ammonium sulfate precipitation techniques is more resource-intensive but provides more accurate results than calculated methods 5

Factors Affecting Calculation Accuracy

• The quality of total testosterone and SHBG measurements significantly impacts the accuracy of calculated free testosterone 5, 1
• Mass spectrometry methods (LC-MS/MS) for measuring total testosterone provide superior sensitivity and specificity compared to direct immunoassays when calculating free testosterone 1
• Different calculation methods yield varying results, with mean biases ranging from 5.8% to 56.0% between different equations 6
• The Vermeulen equation has shown good correlation with other validated methods (r = 0.98 with Sodergard equation) 7

Clinical Implications

• When evaluating testosterone status in men with borderline total testosterone (7.5-12 nmol/L), calculated indices of free testosterone are more useful than total testosterone alone 8
• For routine clinical use, a fixed albumin value of 4.3 g/dL is acceptable for most cFT evaluations 2
• The accuracy of calculated free testosterone is contingent upon the assay method used to measure total testosterone and SHBG 5
• When SHBG levels are within normal ranges (above 30 nmol/L), the Vermeulen equation provides clinically useful results for assessing androgen status 1, 2

Common Pitfalls and Limitations

• Using calculated free testosterone methods when SHBG is below 30 nmol/L can lead to inaccurate results 1, 2
• The Free Androgen Index is not recommended for use in men because of inaccuracy at extremes of SHBG concentration 3
• Different calculation algorithms can produce significantly different results, requiring validation in local settings 7
• Morning serum samples should be used for testosterone assessment due to significant diurnal variation in levels 4
• Some calculation methods show confounding by SHBG concentrations, potentially introducing bias 7