Following a cross-sectional study, researchers working in Bologna, Italy, have determined that 9 in 10 women with polycystic ovary syndrome (PCOS) are hyperandrogenic when a steroid profile is included in the definition.
“We confirm here that the combined use of a triad of parameters, including total testosterone, androstenedione, and free androgen index, can accurately define hyperandrogenemia in the majority (90%) of PCOS patients and that their blood concentrations tend to increase with the severity of the phenotype,” wrote the researchers. “In addition, this steroid profile has been proved to help in defining the different phenotypes, independently of the association with hirsutism, with high sensitivity in categorizing the heterogeneous presentation of hyperandrogenemia in PCOS.”
Researchers recruited 156 consecutive white women of reproductive age being treated for PCOS at the Endocrinology Unit of the Sant’Orsola-Malpighi Hospital and 141 age-matched controls.
- All women with PCOS were initially categorized into 3 phenotypes according to the combination of the 3 major Rotterdam’s criteria:
- Patients with all criteria (31%) went into the combination of hirsutism and high testosterone (HA) plus oligo-amenorrhea (OA) plus polycystic ovarian morphology (PCOm) group
- Patients with HA and OA (42%) went into the HA plus OA group
- Patients with OA and PCOm (28%) went into the OA plus PCOm group
No patients had the HA plus PCOm phenotype.
Researchers compared anthropometry, metabolic parameters and steroids between the 3 PCOS phenotypes and between each phenotype and controls.
Results showed that women in the HA plus OA (P=.01) and OA plus HA plus PCOm (P=.001) groups had significantly higher testosterone levels compared with control group patients, with an increasing trend according to phenotype severity (P=.001). At variance, androstenedione and free androgen index levels increased progressively in the 3 PCOS phenotypes compared with the control group. The OA plus HA plus PCOm phenotype had highest androstenedione and free androgen index levels (P=.001).
There was no difference in 17-OHP and 11-deoxycortisol levels between the control group and the 3 phenotype groups.
Among the women with PCOS, 56% had high androstenedione, 73% within the HA plus OA plus PCOm group, 58% within the HA plus OA phenotype, and 33% within the OA plus PCOm phenotype. Overall, including androstenedione in the diagnosis of biochemical hyperandrogenism led to a prevalence of hyperandrogenic PCOS of 81%.