Key clinical features
PCOS has serious clinical sequelae, including psychological problems (poor self-esteem, anxiety), reproductive manifestations (hirsutism, infertility and pregnancy complications) and metabolic implications (insulin resistance, metabolic syndrome, prediabetes, type 2 diabetes and potentially cardiovascular disease) (table 3).
Given the heterogeneous nature of PCOS (see Prevalence) and the spectrum of clinical features, presentation can vary across the life cycle. Essentially PCOS is a chronic condition with manifestations that begin most commonly in adolescence with oligomenorrhoea/amenorrhoea and transition into problems including infertility and metabolic complications over time.
However, when combined with obesity, metabolic features present earlier in life, with diabetes not uncommonly diagnosed on screening in adolescents and women in their 20s and 30s.Ovarian dysfunction/menstrual cyclicity
Ovarian dysfunction usually manifests as oligomenorrhoea/amenorrhoea resulting from chronic oligoovulation/anovulation. Irregular cycles of >35 or <21 days are considered oligomenorrhoea. However, prolonged anovulation can lead to dysfunctional uterine bleeding, which may mimic more regular menstrual cycles. In these patients, oligoanovulation can be confirmed by measurement of a luteal phase day 22-24) progesterone23.
Most patients with PCOS have ovarian dysfunction, although women with regular menstrual cycles can also be diagnosed with PCOS, based on other diagnostic criteria (see Prevalence). Among women with PCOS, 70-80 per cent have oligomenorrhoea or amenorrhoea. Among women with oligomenorrhoea, 80-90 per cent will be diagnosed with PCOS, while only 40 per cent of women with amenorrhoea will be diagnosed with PCOS, as hypothalamic dysfunction is a more common cause.
On history, oligomenorrhoea usually occurs in adolescence, or (if onset is later in life) it often follows weight gain or physical inactivity. Recent guidelines recommend that in adolescent women (<18 years) after two years of irregular cycles following the onset of menarche, assessment for PCOS should be considered24. Menstrual irregularity is often subsequently camouflaged by use of the oral contraceptive pill (OCP). When the OCP is stopped, underlying irregular cyclicity recurs. To address this issue, recent guidelines have provided suggestions if the OCP is considered in adolescents (35 or <21 days), after the onset of menarche, PCOS should be considered before commencement of the OCP. Where the OCP has already been commenced, when girls are not sexually active, it can be withdrawn for at least three months before hormonal assessments for PCOS are undertaken. The risk of unplanned pregnancy needs to considered and weighed up against potential benefits of early diagnosis. Contraception should be otherwise managed during this time24. Menorrhagia can occur in the setting of unopposed oestrogen, with endometrial hyperplasia, often exacerbated by elevated oestrogen levels in obesity. Women with PCOS are at increased risk of developing endometrial cancer25. While inadequate research exists, it is generally recommended that at least four cycles a year are important to protect against endometrial hyperplasia.
Infertility
PCOS is the most common cause of anovulatory infertility. It accounts for 90-95 per cent of women attending infertility clinics with anovulation. However, 60 per cent of women with PCOS are fertile (defined as the ability to conceive within 12 months), although time to conceive is often increased.
Anovulation is the most common cause of infertility in PCOS. A sub-group of women have impaired oocyte development following controlled ovarian stimulation for in vitro fertilisation and a higher rate of miscarriage which is exacerbated by higher IR and hyperandrogenaemia (reviewed in23).
In those with PCOS and infertility, 90% are overweight. Obesity also independently exacerbates infertility, reducing efficacy of infertility treatment and inducing a greater risk of miscarriage. Australian guidelines suggest that pharmacological ovulation induction should not be recommended for first line therapy in women with PCOS who are morbidly obese (BMI > 35 kg/m2) until weight loss has occurred either through diet, exercise, bariatric surgery or other appropriate means24. Regardless, the risks of pregnancy in overweight women are considerable for both mother and baby. Age-related infertility also exacerbates infertility in these women, and timely planning of families may warrant discussion.
Complications in pregnancy and the neonatal period
Women with PCOS are at increased risk of gestational diabetes26, pregnancy-induced hypertension, pre-eclampsia and preterm birth. Neonatal complications include increased risk of mortality and neonatal intensive care admission26.Hyperandrogenism
The clinical and/or biochemical signs of androgen excess in PCOS result primarily from increased synthesis and release of ovarian androgens. Elevated LH and insulin levels synergistically increase stimulation of theca cells to synthesise androgens (see Prevalence). Insulin resistance leads to hyperinsulinaemia, reduces SHBG and raises free circulating testosterone levels. Hyperandrogenism and hyperinsulinaemia impair follicle development in the ovary.
Clinical hyperandrogenism primarily includes hirsutism, acne and male-pattern alopecia. The clinical signs of androgen excess result from the increased exposure to dihydrotestosterone, synthesised in the skin from circulating androgens and affecting hair follicles.
Hirsutism is defined in females as male-type terminal hair growth and distribution. PCOS is a common cause of hirsutism; about 60-80 per cent of women with PCOS have hirsutism, although this varies with race and degree of obesity. Hirsutism should be assessed with a standardised scoring system (Ferriman-Gallwey score [in Caucasian women a score of 8 or higher indicates androgen excess])27. However Asian women show less clinical hyperandrogenism and South East Asian and Mediterranean populations manifest significant clinical hyperandrogenism. Acne affects one-third of patients.
Male-pattern hair loss (androgenic alopecia) is not frequently seen in PCOS cases, as it usually requires a familial predisposition. Other features of hyperandrogenism include virilisation, which, especially if present as clitoromegaly and rapid in onset, requires exclusion of other causes, including adrenal or ovarian androgen-secreting tumours.
Biochemical hyperandrogenism is present in most patients with PCOS. Measurement of biochemical androgens in PCOS is limited by poor accuracy and reproducibility of assays, which are designed for significantly higher levels of androgens in males. In general it is recommended that calculated free testosterone, calculated bioavailable testosterone or free androgen index, which is derived by the lab from SHBG and total testosterone measurements, are performed. Androgen testing should be performed in the follicular phase in cycling women and after withdrawal of the oral contraceptive pill for at least three months.
Tests for dehydroepiandrosterone-sulphate (DHEAS) and androstenedione could be considered as second-line investigations to exclude other causes of hyperandrogenism in PCOS24. DHEAS may be mildly raised in PCOS but markedly raised in the setting of androgen secreting tumours, however a clinical history of rapid virilisation is more useful23. Androstenediones may be mildly elevated in PCOS but marked elevations are more indicative of non-classical adrenal hyperplasia24.
Late onset congenital adrenal hyperplasia should be excluded before PCOS is diagnosed. Serum 17-hydroxyprogesterone should be measured in the follicular phase if possible.
Metabolic features: PCOS, impaired glucose tolerance and type 2 diabetes mellitus
It is now recognised that the majority of PCOS women have increased insulin resistance compared to non-PCOS women matched for BMI and body fat distribution28-31. Insulin resistance is clinically relevant as it drives hyperandrogenism, underpins reproductive features including infertility and is a key factor in the metabolic features of PCOS28,29. Insulin resistance appears to affect 60-80 per cent of women with PCOS.
The cause of insulin resistance is complex and multifactorial, with genetic and environmental contributors. While these include obesity, lean women with PCOS also have abnormalities of insulin secretion and action compared with weight-matched controls30. However, insulin resistance is likely most severe in those diagnosed by NIH criteria and in those affected by obesity14. Women with more severe IR in PCOS present with more severe clinical features32.
Existing research into insulin resistance in PCOS is limited to humans, with no ideal animal models. Specific abnormalities of insulin metabolism identified in PCOS include reductions in secretion33,34, reduced hepatic extraction34, impaired suppression of hepatic gluconeogenesis30 and abnormalities in skeletal muscle and adipose tissue insulin signalling35,36.Insulin resistance appears to contribute to the reproductive manifestations of PCOS. Interestingly, there is a paradoxical expression of insulin resistance in PCOS whereby insulin-stimulated androgen production persists, while its role in glucose metabolism is impaired. Therefore, insulin resistance in PCOS results in hyperinsulinaemia with its associated diverse and complex effects on regulating lipid metabolism, protein synthesis and modulation of androgen production.
Insulin resistance is an independent predictor of impaired glucose tolerance and diabetes in the general population [reviewed in 37].A subgroup of women with PCOS and insulin resistance also develops insufficient pancreatic insulin output or beta-cell failure. In this setting, insulin output cannot overcome insulin resistance, and hyperglycaemia develops. Women with PCOS are at increased risk of developing impaired glucose tolerance (IGT) and type 2 diabetes (with prevalence rates of 31.3 and 7.5 per cent, respectively, compared with 14 per cent for IGT and zero for type 2 diabetes in age- and weight matched non-PCOS control women)38. A recent systematic review and meta-analysis found that women with PCOS exhibited an increased prevalence of impaired glucose tolerance [odds ratio (OR) 2.52, BMI-matched studies OR 2.56], diabetes (OR 4.44, BMI-matched studies OR 4.06) and metabolic syndrome (OR 2.89, BMI-matched studies OR 2.12)39.Adolescents with PCOS also have high rates of IGT and type 2 diabetes40.There is also emerging evidence that women with PCOS have a greater chance of developing gestational diabetes, with a recent meta-analysis reporting an odds ratio of 2.9426. These elevated risks occur both independently of obesity, and as a result of obesity. Furthermore, the International Diabetes Federation has identified PCOS as a significant non-modifiable risk factor for type 2 diabetes.
Women with PCOS also appear more likely to develop abnormal glucose metabolism at a younger age and may demonstrate a more rapid conversion from IGT to type 2 diabetes. The rate of conversion from IGT to type 2 diabetes in the general population was estimated in the AusDiab (Australian Diabetes, Obesity and Lifestyle) study as 2.9 per cent per year for females. An Australian study in PCOS has reported a substantially higher conversion rate (8.7 per cent per year over 6.2 years)41 but this has not been uniformly reported. Prospective trials show a 5-10 fold increase in progression from impaired glucose tolerance to diabetes in PCOS38,42. One study reported a fivefold increased risk of developing diabetes over a period of eight years in obese women with PCOS, compared with the risk calculated using age-adjusted obese controls43. Legro et al. reported an OR of 2.4 over a two to three year follow-up period for conversion from normal glucose tolerance to impaired glucose tolerance / diabetes in PCOS and controls44.There are few adequately powered studies assessing the natural history of IGT, type 2 diabetes and cardiovascular disease in PCOS. Because of this, there is a need for a definitive, large, long-term prospective study assessing these risks in women with PCOS.
It is increasingly clear that IGT is also a clinically relevant state, the identification of which can lead to interventions that improve long-term outcomes. IGT has been found to increase the risk of cardiovascular disease, mortality and progression to diabetes in general populations. Recent AusDiab data found a mortality rate of 5.5 per cent over five years for people with IGT, compared to 1.9 per cent for those with normal glucose tolerance.
Furthermore, lifestyle intervention together with use of metformin can prevent IGT progression to type 2 diabetes (although use of metformin in this way would be off-label), strengthening the argument for early detection of IGT, especially in this group of women who are at high risk of type 2 diabetes.
In PCOS:
- Insulin resistance is increased
- Metabolic profiles are adverse
- IGT is more common
- Type 2 diabetes risk is increased 4-7 fold
- Cardiovascular risk factors are elevated
- Cardiovascular disease risk is likely to be increased
We therefore postulate that detecting and treating IGT is particularly clinically relevant for women with PCOS, to identify those at high risk for progression to type 2 diabetes and for aggressive cardiovascular risk factor reduction.
Although IGT and type 2 diabetes are clinically relevant and common in PCOS, there are no screening guidelines specifically to detect IGT or diabetes in this population. However, guidelines have been designed to screen for diabetes in the general population. These guidelines frequently use fasting glucose as a screening test and only progress to an oral glucose tolerance test (OGTT) if the fasting level is abnormal, yet impaired fasting glucose is a poor predictor of IGT, especially in PCOS. Furthermore, in general populations impaired fasting glucose and IGT comprise two distinct subgroups with only partial overlap.
Likewise, in PCOS, fasting plasma glucose levels >6.1mmol/L are also a poor predictor of IGT, leading to recommendations by our group and others, including the Rotterdam PCOS Consensus Workshop Group, to perform an OGTT in obese women with PCOS. In our work to date, testing fasting glucose levels misses 80 per cent of IGT in women with PCOS and we strongly endorse screening with an OGTT in overweight/obese women with PCOS every 1-2 years.
As a final point here, emerging data show an increased risk of metabolic syndrome, IGT and diabetes in first-degree family members of women with PCOS and in relatives with other risk factors for diabetes (e.g. age, family history, overweight, high-risk ethnic group). Therefore screening for cardiometabolic abnormalities may be warranted in this group.
Cardiovascular disease risk
Insulin resistance is an independent predictor of cardiovascular risk in the general population37. Along with insulin resistance, metabolic syndrome, IGT and diabetes, women with PCOS also have an increased prevalence of traditional risk factors (dyslipidaemia, hypertension) and novel risk factors (increased homocysteine levels, inflammation, oxidative stress, leucocyte counts and impaired fibrinolysis) for cardiovascular disease.
A recent meta analysis confirms that women with PCOS have higher trigycerides, LDL-cholesterol, non-HDL cholesterol and lower HDL cholesterol than non-PCOS women, regardless of BMI45. Jean Hailes research has demonstrated that early clinical and subclinical markers of atherosclerosis (endothelial dysfunction, impaired pulsewave velocity, increased carotid intima-media wall thickness, presence of carotid plaque and increased coronary artery calcification) are more advanced in PCOS and are associated with insulin resistance and obesity.
There is a lack of prospective long term studies in PCOS to appropriately address the issue of cardiovascular disease risk. However, the higher rates of metabolic syndrome and increased macrovascular disease in people with type 2 diabetes would be expected to lead to increased cardiovascular disease. The AusDiab study reported that 65 per cent of deaths due to cardiovascular disease occurred in subjects with type 2 diabetes or impaired fasting glucose (IFG) or IGT; a 50-60 per cent greater cardiovascular mortality risk for subjects with IFG or IGT; and a twofold increased risk for subjects with type 2 diabetes46. As women with PCOS are a population at high risk for developing type 2 diabetes, they are also likely to have increased cardiovascular and metabolic mortality and morbidity.
Recent epidemiological studies have strongly suggested PCOS women are at increased CVD risk. The recent Consensus Statement by the Androgen Excess and Polycystic Ovary Syndrome Society concluded that the evidence for increased cardiovascular morbidity and mortality is inconclusive, yet suggestive, and that the current epidemiological data suggests more frequent cardiovascular disease in NIH diagnosed PCOS47,48. This was supported in a new meta-analysis in PCOS showing a two-fold increase in cardiovascular disease and stroke in PCOS women relative to women without PCOS which remained after adjustment for BMI49.
Psychosocial factors
Most research has focused on the biological and physiological aspects of the syndrome. However, the consequent impact of physical symptoms, infertility and long-term health-related concerns on mood and psychological wellbeing of women with PCOS is also likely to be considerable.
Although research in this area is inadequate, it suggests that challenges to feminine identity and body image due to obesity, acne and excess hair, compromise quality of life in women with PCOS. Limited studies to date have reported that women who have PCOS are more prone to depression (28-64 per cent), anxiety (34-57 per cent), low self-esteem, negative body image disordered eating (21 per cent) and psychosexual dysfunction and have reduced quality of life, although further research is needed (reviewed in 24).
The other critical aspect of psychosocial impact in PCOS is the negative impact of mood disturbance, poor self-esteem and reduced psychological wellbeing on readiness to change and on ability to implement and sustain successful lifestyle changes that are so critical in this condition.
Hence these issues need to be explored and addressed as a critical part of PCOS assessment and management. Recent Australian guidelines suggest that screening for depression, anxiety, body image, disordered eating and psychosexual dysfunction should be performed in women with PCOS24.
This article first appeared in Australian Doctor - How to treat on 29 August 2008 and has been reproduced here with permission.
Australian Doctor - How to treat: Polycystic ovary syndrome 437.44 Kb
Content updated 7 September 2011





