Objective: To determine whether ovarian polycystic changes estimated by transvaginal sonography correlate with clinical and endocrine findings associated with the polycystic ovarian syndrome.
Design: Prospective study.
Setting: Reproductive Endocrinology Unit in the Department of Obstetrics and Gynecology at a teaching hospital.
Participants: Ninety-five consecutive patients suffering from oligoamenorrhea entering an infertility treatment program.
Interventions: All women were examined by transvaginal sonography, assessed for body mass index and hirsutism. Blood withdrawal was performed for hormone estimates.
Main outcome measures: Ovarian follicle number, volume, and stroma echogenicity. Estimates of immunoreactive and bioactive luteinizing hormone (LH), testosterone (T), free T, insulin levels, and insulin resistance index were performed.
Results: Hirsutism was present in 63% of the patients and correlated with the number of follicles, ovarian volume, and stroma echogenicity. Follicle number, ovarian volume, and stroma echogenicity values were significantly correlated with immunoreactive LH, bioactive LH, and T levels. Both T and immunoreactive LH were independently correlated with ovarian structure sonography parameters. Insulin and insulin resistance correlated with ovarian volume and stroma echogenicity. Insulin resistance was of significant additional predictive value of ovarian volume and amount of stroma. Evaluating the predictive value of immunoreactive LH and T together in regard to all sonography parameters, only T levels were statistically significant predictors of increase in follicle number, ovarian volume, and stroma amount.
Conclusions: Although immunoreactive LH and bioactive LH correlated with all ovarian parameters, the effect of androgens on ovarian changes appeared to be independent from LH. This further substantiates the apparent cardinal role of androgens in the genesis of polycystic ovaries. Besides, transvaginal sonography assessment of ovaries is a valuable additional tool for the diagnosis of PCOS. In this regard, insulin resistance is of additional predictive value for ovarian volume and stroma echogenicity.