Pathophysiology of polycystic ovary syndrome
Until now, the underlying cause of PCOS remained elusive. However, there are a number of proposed mechanisms which involve hyperandrogenism, hyperinsulinism, and high levels of anti- Mullerian hormone (AMH) associated with anovulation as the pathogenesis of PCOS.
Androgen excess
Hyperandrogenism can be a cause of PCOS but it remains controversial and difficult to prove in humans (5). Clinical studies demonstrated that androgen levels such as total testosterone and androstenedione were consistently elevated in women with PCOS when compared to controls (6, 7). Giving the androgen receptor antagonist flutamide to some women with PCOS restored menstrual regularity and ovulation in these women (8). This study seemed to suggest that androgen receptor-mediated androgen actions could be involved in the pathogenesis of PCOS resulting in the phenotype we recognize in women with PCOS such as irregular periods, polycystic ovarian morphology, or signs of androgen excess. Additionally, an association between short CAG repeat length in the androgen receptor and the subset of anovulatory patients with low serum androgens suggests that the pathogenic mechanism of polycystic ovaries in these patients could be due to the increased intrinsic androgenic activity associated with short androgen receptor alleles (9). Walters reviewed the literature and concluded that rodent, sheep, and primate PCOS models had clearly demonstrated that the reproductive, endocrine, and metabolic characteristics of human PCOS (anovulation, polycystic ovaries, increased anthral follicular count, luteinizing hormone elevation, insulin resistance, dyslipidaemia, obesity) could be consistently induced by androgen excess (5). All in all, the data suggest that androgen excess involving the androgen receptor-androgens pathway is intrinsically linked to PCOS.
Hyperinsulinism
De Leo and colleagues had suggested that insulin resistance and hyperinsulinism could have a central role in the pathogenesis of PCOS (10).
Human ovaries had specific receptors for insulin (11, 12) and studies on thecal cells also demonstrated that insulin acted as a co-gonadotropin in steroidogenesis (14) and stimulated proliferation of thecal cells (15) to increase luteinizing hormone-stimulated androgen secretion (16-18) and to increase P450c17 mRNA levels (19) which upregulated luteinizing hormone receptors (20) and ovarian insulin-like growth factor 1 receptors (13). Hyperinsulinaemia therefore increased androgen levels by stimulating ovarian steroidogenesis and inhibiting insulin growth factor binding protein and sex hormone-binding globulin (SHBG) synthesis and secretion (10). Based on the review by De Leo and colleagues, only metformin and the new thiazolidinediones (rosiglitazone and pioglitazone) should be recommended in clinical practice as the balance of opinion seems to favour beneficial effects of insulin-lowering agents on insulin sensitivity, hyperandrogenaemia, menstrual irregularity, and metabolic disorders in a large subset of affected women with PCOS (10). Another study also demonstrated that in women with PCOS, treatment with metformin is effective in the lowering of hyperinsulinaemia, hyperandrogenaemia, and in many women with PCOS, improves the menstrual pattern, but
has no effect on hirsutism (21). A recent review by Lautatzis and colleagues also affirmed the efficacy and safety of metformin use throughout pregnancy in women with PCOS as this intervention reduced the rates of early pregnancy loss, preterm labour, and protected against fetal growth restriction. Also, there were no demonstrable teratogenic effects, intrauterine deaths, or developmental delays with the use of metformin in pregnancy (22). Based on the in vitro and clinical studies to date, hyperinsulinism can potentially be an aetiological agent for the PCOS phenotype.
Anti-Mullerian hormone
PCOS is a heterogeneous disorder affecting ovarian function primarily, and has features of multiple small follicles growing but which arrest and result in anovulation.
Interestingly, AMH is unique and specific to the human ovary and its expression is restricted to the granulosa cells of the ovary in women. AMH begins to be produced at around the 25th week of gestation and continues until menopause (23, 24). AMH is also expressed at all steps of folliculogenesis and is initiated as soon as primordial follicles are recruited to grow into small preantral follicles with its highest expression being observed in preantral and small antral follicles. Thereafter, AMH expression decreases with the selection of follicles for dominance and becomes no longer expressed during the follicle-stimulating hormonedependent stages of follicular growth (except in the cumulus cells of preovulatory follicles) and in atretic follicles (25, 26). Dumont and colleagues performed a comprehensive review of AMH and its association with PCOS and concluded that in PCOS there were multiple levels of abnormalities in folliculogenesis attributable to the actions by high levels of AMH resulting in an increased number of small growing follicles and inhibition of the terminal follicular growth. This will lead to a lack of selection of the dominant follicle known as ‘follicle arrest’. Additionally, a possible follicular apoptosis defect can result in the excess of growing follicles (27).Henceforth, the most striking reproductive disorder based on the proposed mechanisms results in the clinical presentation of anovulation as irregular (i.e. long intervals between menstrual cycles) or absent periods. This has been demonstrated in a recent study by Zhu and colleagues showing the association between higher AMH levels and longer menstrual cycle length of more than 35 days (28). Due to the anovulatory cycles, women with PCOS may also encounter issues with other endocrine disturbances secondary to the androgen excess and hyperinsulinism. An attempt to look at genetic variants in the AMH signal transduction pathway demonstrated population differences; however, this does not appear to have significant effects on ovarian, endocrine, and metabolic parameters and reproductive outcomes (29).