Chapter 20 Polycystic ovarian syndrome
CLASSIFICATION AND AETIOLOGY
Polycystic ovarian syndrome (PCOS) is a term to describe a constellation of clinical and biochemical features. For many of these the aetiology remains poorly understood. Several factors also preclude difficulties in diagnosis of PCOS, including a heterogeneous range of symptoms that can change over time and the lack of a precise and uniform consensus on diagnosis. In 2003 a consensus workshop indicated PCOS to be present if two out of three criteria are met: oligoovulation and/or anovulation, excess androgen activity (as determined by elevated free androgen index) and polycystic ovaries (by gynaecological ultrasound); and other endocrine disorders such as hyperprolactinaemia are excluded.1 Elevated fasting insulin or high insulin levels in glucose tolerance tests may also be used to suggest diagnosis. Other blood tests may be suggestive but not diagnostic, for example if the LH:FSH ratio is greater than 1:1, or if there are low levels of sex hormone binding globulin. The presence of ovarian cysts does not automatically imply a diagnosis of PCOS. The prevalence of PCOS is thought to be between 5 and 10% of women and is one of the main causes of infertility in Western women.2,3
The symptoms of PCOS usually appear upon menarche, and are associated with early puberty brought about by early secretion of androgens.4 This may also be associated with low birth weight. However, the condition can develop a considerable time after menarche in the presence of other environmental factors such as weight gain and subsequent insulin resistance.
Increased ovarian androgen biosynthesis in the polycystic ovary syndrome results from abnormalities at all levels of the hypothalamic–pituitary–ovarian (HPO) axis. Androgen excess in women with PCOS may be of either ovarian or adrenal origin. It is also postulated that insulin may induce overactivity of 11b-hydroxysteroid dehydrogenase, resulting in excessive adrenal androgen production.5 Androgens may be converted to oestrone in fatty tissue, causing blood oestrone and ultimately stimulating LH production, which triggers ovarian androgen production. Increased frequency of luteinising hormone (LH) pulses in PCOS may result from an increased frequency of hypothalamic gonadotrophin-releasing hormone (GnRH) pulses, resulting in higher production of LH compared with follicle-stimulating hormone (FSH).The increase in pituitary secretion of LH can lead to an increase in androgen production by ovarian theca cells. Increased efficiency in the conversion of androgenic precursors in theca cells leads to enhanced production of androstenedione, which can then be converted by 17b-hydroxysteroid dehydrogenase (17bOHSD) to form testosterone or aromatised by the aromatase enzyme to form oestrone, which can be further converted to oestradiol by 17bOHSD (see Figure 20.1).5
Insulin acts synergistically with LH to enhance androgen production. Insulin also inhibits hepatic synthesis of sex hormone-binding globulin (which ordinarily binds to testosterone) and therefore increases the proportion of testosterone that is biologically available. Testosterone inhibits and oestrogen stimulates hepatic synthesis of sex hormone-binding globulin.5
Polycystic ovaries in PCOS develop when the ovaries are stimulated to produce excessive amounts of androgens—particularly testosterone—through the release of excessive luteinising hormone (LH) by the anterior pituitary gland or through high levels of insulin in the blood.6 This causes the follicle to begin maturation but the lack of LH surge results in anovulation, meaning the ovum does not release, and ultimately a cyst is formed (see Figure 20.2).
RISK FACTORS
A majority of, though not all, patients in Western settings with PCOS have insulin resistance and are often overweight.7 Elevated insulin levels contribute to or cause the abnormalities seen in the hypothalamic–pituitary–ovarian (HPO) axis that lead to PCOS. Hyperinsulinaemia may increase GnRH pulse frequency, LH dominance over FSH, increased ovarian androgen production, decreased follicular maturation, and decreased SHBG binding—all of which can lead to the development of PCOS.5 Insulin resistance is a common finding among patients of normal weight as well as overweight patients.
It is easy to view the typical PCOS patient with insulin hypersensitivity as the conventional overweight type. However, insulin resistance is also common in lean women with PCOS.8 It is also easy to view PCOS as a primarily androgen-dependent disorder; however, oestrogen dominance is also commonly present in women with PCOS. Although weight loss is generally associated with good clinical effect and being overweight is common in women with PCOS, it should never be assumed that women with PCOS will always be overweight. This trend should be noted to prevent the misdiagnosis of the condition in women who are not overweight.
Thyroid problems may make PCOS symptoms worse9 and women with PCOS also have a high prevalence of autoimmune thyroid conditions.10 Thyroid function should therefore also be checked in patients with PCOS (see Chapter 17 on thyroid abnormalities).
Liver support may also be required in women with PCOS. Approximately 30% of women with PCOS have raised liver enzymes, and diabetes and insulin resistance also increase the risk of non-alcoholic liver disease.11,12 Hyperinsulinaemia may inhibit the production of SHBG in the liver.
CONVENTIONAL TREATMENT
Due to the functional nature of diagnosis, conventional treatment in PCOS is generally focused on a number of goals. These may include reducing hyperinsulinaemia; restoring normal menstruation, reproductive function and fertility; and reducing associated symptoms such as hirsutism.13
If pregnancy is desired, assisted reproductive techniques, including radical drug therapy with clomiphene or a similar agent, or in vitro fertilisation techniques, are also commonly prescribed (see Chapter 31 on fertility, preconception care and pregnancy).13
NATUROPATHIC DIAGNOSIS TECHNIQUES
Charting
Due to the often significant timeframes encountered when treating patients with PCOS, tools such as charting may be employed to observe hormonal status changes over time. Menstrual cycle charting based on basal body temperature (BBT) has been traditionally used in naturopathic practice to ascertain changes in levels of hormone levels in reproductive females (see Figure 20.3). It is thought to reflect the menstrual cycle in two ways: (1) within 1–2 days before LH surge there should be a low point in BBT; and (2) following ovulation women should generally see a sustained increase in BBT of between 0.2 and 0.5°C.14 This rise is most likely due to the thermogenic effect of a metabolite of progesterone.
It should be noted that focusing on individual components of charting is very often inadequate.14 However, when integrated they can form a useful clinical tool that provides a broad overview of reproductive function and actively involves patients in their treatment. Despite the fact that more specific and accurate investigations currently exist, charting may still have a role to play in clinical practice as it is inexpensive, non-invasive and generally reliable. It also actively involves the patient in the therapeutic and diagnostic process. However, interpretation can be subjective and difficult and in more complex cases other investigative techniques may be more appropriate.
Consistently low temperature (under 36.5°C) may indicate a hypothyroid condition, which may play a role in poor reproductive function (see Chapter 17 on thyroid abnormalities).
Progesterone
Adequate progesterone levels will result in a sustained, noticeable and prompt rise in BBT. A small or unsustained rise with dips in temperature will be indicative of poor progesterone levels. However, temperature levels alone do not confirm appropriate levels of progesterone. The luteal phase needs to be at least 11 days to indicate adequate progesterone levels. Progesterone levels need to peak approximately 7 days after ovulation to ensure adequate uterine lining for implantation.
KEY TREATMENT PROTOCOLS
Follicle stimulating hormone and luteinising hormone regulation
Some herbal medicines may affect levels of FSH and LH and may therefore proffer some benefit in the treatment of PCOS. Despite its status as a phytoestrogen, Cimicifuga racemosa does not seem to affect the release of prolactin and FSH, though it does reduce LH, in the limited research currently conducted.15 This central effect is thought to be due to its role on dopaminergic regulation of reproductive hormones rather than its effects on oestrogen receptors.16,17 The British Herbal Pharmacopeia lists C. racemosa’s main actions as being useful in ovarian dysfunction and ovarian insufficiency.18
Vitex agnus-castus has had conflicting results, with some studies showing no change in FSH or LH, and another suggesting increased LH release.19–21 V. agnus-castus is thought to have antiandrogenic properties. Humulus lupulus also reduces LH with continued use and may therefore be useful to reduce androgens in PCOS.22 A review of soy studies suggests that soy consumption has no effects on FSH or TSH.23
Unpublished (though publicly available) data suggest that supplementation with the herb Tribulus terrestris for 3 months may normalise ovulation and result in pregnancy in women with endocrine infertility.24 Mentha spicata (via tea therapy—2 cups per day for 5 days) use has been associated with increases in LH, FSH and oestrogen levels in women with PCOS.25
Glycyrrhiza glabra is a commonly used herb in the treatment of PCOS. Though trials for its stand-alone treatment in PCOS are lacking, there exists a theoretical basis behind its use. G. glabra has been demonstrated to reduce testosterone levels in healthy women.26 Various forms of Glycyrrhiza has been used in a combination product (as the key ingredient with Paeonia lactiflora) in the treatment of PCOS in some small trials that showed reduction in FSH:LH ratio, ovarian testosterone production and improvements in ovulation.27,28 However, the situation in the combination product has been confused further by the fact that while this research has focused on G. uralensis clinical application is still dominated by G. glabra. The clinical effects of these minor differences are unknown. G.racemoza may also exert further potent phytoestrogenic activity independent of these effects.29 Glycyrrhiza glabra is also demonstrated to reduce body fat mass in normal weight subjects.30 A Paeonia lactiflora and Glycyrrhiza glabra combination has been found to have numerous effects on PCOS, including regulating FSH:LH ratios (possibly through stimulation of pituitary dopamine receptors),27 lowering testosterone levels and improving oestradiol to testosterone ratios.28,31
Exogenous melatonin has been demonstrated to enhance LH secretion, LH pulse amplitude and LH sensitivity to GnRH.32–34 This was thought to be the result of melatonin supplementation mimicking the effects of PCOS—possibly due to the effects of melatonin on increasing cortisol levels. This effect is known to increase in hypo-oestrogenic states.35 However, while most studies have been in postmenopausal or older women, melatonin supplementation in younger women has been associated with return to menstrual regularity, as well as the reduction of LH levels in younger women with high baseline levels, in clinical settings.36,37 This has led to calls for a possible role of melatonin in the treatment of PCOS and, despite its popularity in some streams of naturopathic medicine, it remains unclear at this stage what role it may play; further research is required.38 Other factors known to affect melatonin levels should also be considered (see Chapter 14 on insomnia).
Other factors may also affect LH levels. Various human studies have suggested that inadequate nutrition or short periods of fasting reduces LH pulsing, though not always LH levels.39–41 This is thought to be due to relative increases in cortisol caused by these states.42 Animal studies seem to suggest that increased endotoxin load can reduce plasma LH levels.43
HPO and hypothalamic–pituitary–adrenal (HPA) axis interaction
It is well known that women with PCOS exhibit abnormalities in cortisol metabolism as well as higher levels than controls.44 It is also known that a return to states of normal cortisol from high cortisol level in women with amenorrhoea or menstrual disturbances often precedes the resumption of normal ovarian activity.45 These findings further support the role for stress reduction in regulating LH and FSH function in the treatment of reproductive disorders such as PCOS.
The reproductive axis is inhibited at all levels by various components of the HPA axis. Corticotrophin-releasing hormone (CRH) can either directly or indirectly (through β-endorphin) suppress gonadotrophin-releasing hormone. Glucocorticoids may also exert inhibitory effects by rendering target tissues resistant to reproductive hormones, inhibiting GnRH and LH secretion and inhibiting ovarian oestrogen and progesterone biosynthesis.5 The effects of the HPA axis interaction with the female reproductive system can result in idiopathic or hypothalamic amenorrhoea (for example, that associated with stress, depression, anxiety or eating disorders) in its own right, or result in the hypogonadism associated with Cushing’s syndrome,5 though it may also result in further indirect complication of other disorders of hormonal dysregulation like PCOS.
However, these interactions can also be bidirectional. Corticotrophin-releasing hormone, for example, is regulated to some extent in reproductive tissue by oestrogen. Corticotrophin-releasing hormone is responsible for a number of functions in reproductive tissue (see Table 20.1), and disorders or events (such as chronic stress adaptation) that affect these levels may also have clinically relevant effects on reproductive function. Due to this bidirectional activity, a multifaceted approach to disorders, focusing on regulation of both reproductive and adrenal hormones, may be more successful in conditions such as PCOS rather than targeting one system alone, particularly considering PCOS may often be a disorder of oestrogen excess (via adipose tissue) as well as androgen excess. For this reason, generalised hormone-balancing protocols may also be beneficial in PCOS. Further information on balancing reproductive hormones can be found in Chapter 19 on endometriosis.
REPRODUCTIVE CRH | POTENTIAL PHYSIOLOGICal ROLES | POTENTIAL PATHOGENIC EFFECTS |
---|---|---|
Ovarian CRH | ||
Uterine CRH | Decidualisation | Infertility (↓ secretion) |
Blastocyst implantation | Recurrent spontaneous abortion (↓ secretion) | |
Early maternal tolerance | ||
Placental CRH | Labour | Premature labour (↑ secretion) |
Weight management
Although not all people presenting with PCOS are overweight, in those that are, weight loss is an essential part of PCOS treatment. Not only can realistic weight loss result in dramatic improvement in the condition, but being overweight can also make treatment less effective.47,48 Weight loss also proffers more effectiveness than current medication
for insulin resistance and related disorders.49 Specific individual pharmacotherapy of any kind—including that of dietary and herbal supplements—is generally clinically ineffective in weight loss in the insulin-resistant individual50 and therefore therapy should focus on an integrated approach to weight management.
As little as 2–5% reduction in weight can be enough to improve metabolic and reproductive indices in women with PCOS.51 This modest improvement can restore ovulatory function and improve insulin sensitivity by over 70%.52 A 5–10% reduction in weight can reduce central fat stores by 30% and weight loss also increases SHBG concentration, reduces testosterone concentration and androgenic stimulation of the skin (resulting in reduction in hirsutism), improves menstrual function and conception rates and reduces miscarriage rates in women with PCOS.53–59
High protein diets are typically associated with excellent weight loss results in insulin-resistant and PCOS women.60,61 Although higher-protein or lower-carbohydrate diets are often successful in weight loss in women with PCOS, this weight loss may not automatically equate to improvements in insulin parameters or ovarian function.62 However, Mediterranean-style diets have been associated with both weight loss and improved insulin parameters in both generic insulin resistant and PCOS patients.63,64
An Israeli study comparing three common diets—low-fat, low-carbohydrate and Mediterranean—found that all diets were effective.65 However, the low-carbohydrate and Mediterranean diets appeared to be more effective than the low-fat diet over 2 years. This suggests that dietary interventions may best be individualised to patient needs rather than protocol-driven.
Long-term modest weight loss is far more important in PCOS than acute weight loss. In fact drastic weight loss in women with PCOS may have negative effects on reproductive function.66 Patient compliance may be improved in low-carbohydrate or higher-protein diets compared to low-fat diets. This is evidenced by a systematic review of dietary interventions that suggested that participant attrition was higher in low-fat diet