Abstract
Objectives Hyperandrogenism, a key feature of polycystic ovarian syndrome (PCOS), is caused
by excess androgen secretion, most commonly of ovarian origin. Although the serum
total testosterone (TT) levels have long been used as a traditional measure of hyperandrogenemia
in women with PCOS, it is associated with many fallacies due to the fact that a component
of TT is linked to sex hormone-binding globulin (SHBG). Recent research has discovered
that measuring free testosterone levels and computing the free androgen index (FAI),
which is a ratio of TT and SHBG, are better predictors of androgen excess in PCOS.
The aim of this meta-analysis is to determine the association of FAI in diagnosing
hyperandrogenism and its ability to discriminate PCOS from controls.
Materials and Methods The publicly available databases PubMed, Scopus and Web of Science were searched
using MeSH terms, ‘Polycystic Ovarian Syndrome’ OR ‘PCOS’ OR ‘PCOD’ AND ‘Testosterone’
AND ‘Sex Hormone Binding Globulin’ OR ‘SHBG’ to collect the full-text articles for
the retrieval of related data of case–control and cross-sectional studies. The studies
quality was assessed using the Newcastle–Ottawa scale, and a sub-group analysis and
publication bias between the studies was evaluated by funnel plot.
Statistical Analysis The R program (v4.0.3) and R packages ‘metafor’ and ‘dmetar’ were used for statistical
analyses of quantitative data and the plots were generated using ‘ggplot2’ package
through a comparison of pooled SMD by Egger's linear regression and Beggs-Mazumdar
tests.
Results Twenty-four studies involving 7,847 participants including 3,290 controls and 4,557
PCOS were included in the meta-analysis. The pooled data analysis of the included
studies showed that the PCOS women had higher FAI than controls, with SMD of 1.56
(95%CI 1.08–2.04; p < 0.01). The publication bias was tested using a funnel plot and Egger's regression
asymmetry test, which revealed no risk of publishing bias (p = 0.1727). Additionally, the sub-group meta-analysis of geographic region revealed
that FAI levels were more significant in PCOS subjects of Asia and Europe, compared
with the American region.
Conclusion Overall, this meta-analysis indicates that FAI could be a reliable marker to differentiate
PCOS patients from controls in Asian and European ethnicities.
Keywords free-androgen index - meta-analysis - polycystic ovarian syndrome - testosterone -
sex hormone-binding globulin
