In the Journals Plus

Free androgen index ‘misleading’ indicator of hyperandrogenism in women with low SHBG

Show Citation

February 9, 2018

Brian Keevil
Brian G. Keevil

Among women with a low concentration of sex hormone-binding globulin, the free androgen index, or FAI, used to assess hyperandrogenism may provide a “misleading” measurement of free testosterone when compared with the equilibrium dialysis method, according to findings published in Clinical Endocrinology.

“The FAI is popular in routine biochemistry laboratories because it involves only two measurements and the ratio is simple to calculate and easy to report on laboratory computer systems,” Brian G. Keevil, FRCPath, an honorary lecturer at the Manchester Academic Health Science Center at the University of Manchester, United Kingdom, and colleagues wrote. “It has been reported that the FAI appears to be the better diagnostic marker to distinguish hyperandrogenism in patients with PCOS when compared with total [testosterone]. However, the value of the FAI in men has been questioned, and studies on the use of different methods for calculating free [testosterone] in men show that the free androgen index tends to overestimate calculated free [testosterone] when the SHBG concentration is low.”

Keevil and colleagues analyzed free testosterone and total serum testosterone in 130 women with polycystic ovary syndrome, 53 healthy women and 120 healthy men using equilibrium dialysis and liquid chromatography-tandem mass spectrometry (LC-MS/MS). Calculated free testosterone and FAI were also measured. Calculated free testosterone was also measured in a retrospective cohort of 4,223 women with normal testosterone levels and SHBG concentration less than 30 nmol/L who underwent routine assessments for hyperandrogenism between March 2005 and March 2015.

Researchers found that calculated free testosterone correlated with free serum testosterone measurements, and the ratio of calculated free testosterone to free serum testosterone was stable across all SHBG concentrations in women and men.

In comparing FAI against calculated free testosterone in the retrospective cohort of women, linear regression analysis showed that correlation in the initial data set was poor (R2 = 0.86); however, correlation improved after excluding low SHBG results (R2 = 0.97).

In patients with low SHBG and normal testosterone concentrations (< 1.6 nmol/L), researchers observed a gradual increase in the ratio of FAI to calculated free testosterone as the SHBG concentration decreased. The ratio of FAI to calculated free testosterone varied in these patients from 0.18 to 0.8, showing a four- to fivefold difference in results, according to researchers.

“For a given testosterone concentration of 1.5 nmol/L, the FAI could vary between 5, 7.5, 10, 15 and 25 with SHBG concentrations of 30, 20, 15, 10 and 5 nmol/L, respectively,” the researchers wrote.

The researchers noted that the discrepancy between FAI and the “gold standard” measurement for free testosterone should push researchers to use caution in the use of FAI in women with low SHBG concentrations.

“We shouldn’t be using the free androgen index as a measure of hyperandrogenism in women, and this is especially important in the investigation of PCOS when the SHBG is often low,” Keevil told Endocrine Today. “The free androgen index will overestimate hyperandrogenism when the SHBG is low, and it would be much better to use the calculated free testosterone instead. Calculated free testosterone correlates very well with the gold standard equilibrium dialysis LC-MS/MS method for serum free testosterone.

“An interesting area for androgen research is in the newly rediscovered 11-ketosteroids,” Keevil said. “There is now good evidence that 11-keto testosterone, produced from the adrenal production of 11-hydroxyandrostenedione, is as potent an androgen as testosterone. This could be important in the investigation of PCOS, but we need larger studies.” – by Regina Schaffer

Disclosures: The authors report no relevant financial disclosures.