To contextualise risk, the remaining lifetime probability in women at menopause of a fracture at any one of these sites exceeds that of breast cancer (~12%). Also, the likelihood of a fracture at any of these sites is 40% or more in developed countries (Kanis et al., 2000), a figure close to the probability of coronary heart disease. Risk factors for osteoporosis include low body mass index (<19 kg/m²), smoking, early menopause, family history of maternal hip fracture, long-term systemic corticosteroid use and conditions affecting bone metabolism, especially those causing prolonged immobility. Osteoporosis is most common in white women. People with osteoporosis are at risk of fragility fractures, occurring as a result of mechanical forces which would not ordinarily cause fracture. The clinically relevant outcome in evaluating treatments for osteoporosis is the incidence of fragility fracture as otherwise this condition is asymptomatic and therefore undiagnosed. The most common sites for these fractures are the hip, vertebrae and wrist. In the UK, the combined cost of hospital and social care for patients with a hip fracture amounts to more than £1.73 billion per year (Torgerson et al., 2001). This is very similar to the annual £1.75 billion health care system costs of coronary heart disease costs. The cost of treating all osteoporotic fractures in post-menopausal women has been predicted to increase to more than £2.1 billion by 2020 (Burge, 2001).

Cardiovascular system

Young adult women are protected against the development of hypertension and its deleterious consequences in the cardiovascular system. Levels of low-density lipoprotein cholesterol (LDL-C) and very-low-density lipoprotein cholesterol (VLDL-C) are decreased by oestrogen, and the levels of high-density lipoprotein cholesterol (HDL-C) are increased, thereby giving some protection against atherosclerosis. HDL-C is known to promote cholesterol efflux from macrophages in the arterial wall, thereby reducing atheromatous plaque and conferring a protective effect against heart disease. However, after menopause, this protection is lost and the incidence of high blood pressure and associated cardiovascular disease increases to levels similar to those found in age-matched men.

Oestrogen has direct beneficial effects in the control of blood pressure, possibly via regulating endothelium-mediated control of arteriolar tone. In women deprived of oestrogen, endothelium-dependent vasodilation is impaired. This dysfunction is largely associated with a reduction in nitric oxide availability. Oestrogen increases nitric oxide availability by stimulating endothelial nitric oxide synthase (eNOS). Oestrogen also stimulates the production of other endothelium-derived relaxing factors such as prostacyclin (prostaglandin I₂). Research suggests that the oestrogen receptor (ER)α is important in mediating the vascular effects of oestrogen. Studies using selective ERα agonists are being undertaken. However, many pathways are stimulated by oestrogen receptor activation and the relative importance of these different pathways varies from tissue to tissue.

Miscellaneous

Thinning of the skin, brittle nails, hair loss and generalised aches and pains are also associated with reduced oestrogen levels (Hall and Phillips, 2005). The skin is the largest non-reproductive target on which oestrogen acts. Oestrogen receptors, predominantly of the ERβ type, are widely distributed within the skin. Both types of oestrogen receptor (ERα and ERβ) are expressed within the hair follicle and associated structures. Thus, epidermal thinning, declining terminal collagen content, diminished skin moisture, decreased laxity and impaired wound healing (selective ERα ligands are being investigated for their wound-healing properties) have been reported in post-menopausal women.

In addition, women also show increasing body weight associated with ageing. This weight gain tends to increase or begin near menopause. Body fat redistribution to the abdomen also occurs independent of weight gain. This type of centralised abdominal fat distribution is widely recognised as an independent risk factor for cardiovascular disease in women.

Psychological changes

Older age at menarche and younger age at menopause are associated with poorer cognitive functioning during ageing. Recent studies have demonstrated that sex steroids have a multifarious and complex relationship with the central nervous system (Hogervorst et al., 2009). For example, there may be a positive correlation between increasing parity and improved executive functioning in response to oestrogen. This temporal relationship between oestrogen deprivation and response to exogenous oestrogen was clearly exemplified in the memory study arm of the WHI (Coker et al., 2010) and MWS studies (Hogervorst and Bandelow, 2010) at the turn of the twenty-first century.

Depression is twice as common in women as in men and may increase during times of changing hormonal levels such as at menopause. Some aspects of decreased central nervous system function have been related to falling oestrogen levels. Preclinical studies have shown oestrogen to have several positive effects on central nervous system function. Perhaps the most important is promotion of acetylcholine synthesis and increased synaptogenesis in the hippocampus. ERβ is expressed in this area of the brain as well as the entorhinal cortex and thalamus, areas crucially involved in explicit memory. Human neuroimaging studies have also indicated that oestrogen influences regional cerebral blood flow in women.

Management

Hormone replacement therapy

HRT is a complicated clinical issue requiring an in-depth risk/benefit assessment. The vast amount of study data are often conflicting, and careful analysis is required. Many factors need to be reviewed before it is prescribed. One important factor is age, as data have shown that if a woman aged less than 35 has a hysterectomy and a bilateral oophorectomy, her risk of non-fatal myocardial infarction is nearly eight times that of her age-matched counterpart who has retained her ovaries. Age at time of HRT prescription in relation to menopausal age, that is, number of years of oestrogen deprivation before replacement, is also of importance when considering outcomes. Individual differences in hormone metabolism (both endogenous and exogenous) are also likely to be important as several different cytochrome enzymes metabolise oestrogen and may be affected by inherited polymorphisms. Therefore, some women may produce oestrogenic metabolites possessing considerable oestrogenic activity, whilst others produce metabolites which are relatively non-oestrogenic. Body mass index (BMI) also influences response to HRT, with increased plasma estradiol levels observed in women with higher BMIs.

HRT is effective for symptomatic relief of menopausal symptoms, and its use is justified when symptoms adversely affect quality of life. Current advice is that the lowest effective dose for a particular woman should be used for the shortest period of time. Local oestrogen replacement may be used to reverse the symptoms of urogenital atrophy as it appears to be more effective than systemic therapy. There is no evidence to suggest that local oestrogen treatment is associated with significant risks.

Treatment with HRT should be reviewed at least annually, with alternative therapies considered for the management of osteoporosis. In the treatment of menopausal symptoms, the benefits of short-term HRT outweigh the risks in the majority of women, but in healthy women without symptoms, the risks outweigh the benefits.

Contraindications to the use of HRT include undiagnosed vaginal bleeding in post-menopausal women, the presence of an oestrogen-dependent tumour, liver disease (where liver function tests have failed to return to normal), active thrombophlebitis, and active or recent arterial thromboembolic disease, for example, angina or myocardial infarction. A history of deep vein thrombosis and pulmonary embolism requires careful evaluation before the use of oestrogen therapy. Use in patients with Dubin–Johnson and Rotor syndromes may also be contraindicated.

Oestrogen therapy

Since the symptoms and long-term effects of menopause are due to oestrogen deprivation, the mainstay of HRT is oestrogen. This may be administered orally or parenterally but, in either case, the oestrogens used are naturally occurring and include:

• conjugated equine oestrogen (estrone sulphate 40%, equilin sulphate 60%)

• estradiol valerate.

The use of ‘natural’ oestrogens reduces the risk of the potentially dangerous oestrogenic effects such as raised blood pressure, alteration in coagulation factors and an undesirable lipid profile, which sometimes occur with the more potent synthetic oestrogens used in the oral contraceptive agents. A ‘natural’ oestrogen is defined as one that is normally found in the human female and has a physiological effect. Natural oestrogens are less potent (up to 200 times) than synthetic oestrogens. As they are naturally occurring compounds, the plasma half-life of these oestrogens is similar to that of the ovarian-secreted oestrogens and the duration of action is shorter than the synthetic oestrogens, such as ethinylestradiol, used in many formulations of the contraceptive pill. The plasma ratio of estradiol to estrone is normally about 1:1 to 2:1, and the aim of HRT should be to preserve this ratio.

There are four main routes of administration for oestrogens in HRT:

• oral

• transdermal (patches/gels/cream)

• subcutaneous (implants)

• vaginal (creams and medicated rings).

The use of oral oestrogen therapy, while convenient for the patient, does mean that the oestrogen will be subjected to conversion to estrone by the liver and the gut, thereby altering the estradiol:estrone ratio in favour of the less active oestrogen, estrone. The oral preparations have different metabolic effects due to first-pass hepatic metabolism. Smoking stimulates metabolism of oestrogens by cytochrome P450 and decreases plasma oestrogen levels by 40–70% in oral oestrogen users. Smoking has no significant effect on plasma oestrogen levels in users of transdermal preparations. Oral delivery compared to transdermal delivery (Table 46.2) also has different effects on lipid levels (Vrablik et al., 2008). In addition, orally administered oestrogens undergo first-pass hepatic metabolism, which may result in some reduction in anti-thrombin III, a potent inhibitor of coagulation. Implants and patches show smaller changes in coagulation, platelet function or fibrinolysis.

Table 46.2 Effect of HRT administration route on lipid profile

Oral	Transdermal
↓ Low-density lipoprotein	↓ Low-density lipoprotein
↓ Total cholesterol	↓ Total cholesterol
↑ High-density lipoprotein	↔ High-density lipoprotein
↑ Triglycerides	↓ Triglycerides
↑ Bile cholesterol	↔ Bile cholesterol

More constant levels of oestrogen result from the use of transdermal patches containing estradiol, and these have the added advantage of a more physiological estradiol:estrone ratio (Delmas et al., 1999). However, the adhesive used in these transdermal patches and the alcohol base can cause skin irritation. The patch is applied to the non-hairy skin of the lower body, and care should be taken to ensure that it is placed away from breast tissue. The patch is changed either once or twice a week, thus providing a constant reservoir of estradiol to provide a controlled release into the circulation. Estradiol is also available in a gel formulation, applied daily to the skin over the area of a template (to ensure correct dosage), but this formulation may give erratic absorption. The intranasal preparation, administered as a nasal spray, also avoids hepatic first-pass metabolism.

The oestrogen implant gives a constant level of oestrogen from a few days after insertion for up to 6 months. This formulation maintains the best estradiol:estrone ratio and is a convenient method of administration, requiring repeat implants only every 6 months. However, because the levels of oestrogen are constantly raised, there will be some increase in oestrogen receptor numbers, and this can lead to a recurrence of symptoms of oestrogen deficiency due to the presence of unoccupied oestrogen receptors, even in the presence of normal or even high oestrogen levels. This phenomenon, called tachyphylaxis, results in patients becoming symptomatic and requesting repeat implants earlier and earlier. In such cases, it is unwise to treat with additional oestrogen; the patient should receive counselling and perhaps a change of preparation. The disadvantage of the implant is that, once inserted, it cannot be removed readily and even if it is removed, the oestrogen level will take at least a month to fall. There is also evidence that the uterine endometrium, if present, remains stimulated for some time after removal of the implant.

Both the transdermal and implant preparations avoid the first-pass hepatic effects of oral oestrogens and are less likely to affect liver enzyme systems and clotting factors. Some studies show an increase in the incidence of venous thromboembolism (VTE) in women taking HRT. Therefore, patients who have a history of deep vein thrombosis or pulmonary embolism will need careful guidance, with each woman being considered individually, and the relative risks evaluated. Other risk factors include severe varicose veins, obesity or a family history of deep vein thrombosis (DVT). If HRT is justified in such patients, transdermal preparations are a better alternative than oral preparations.

Vaginal creams containing oestrogen are available but generally fail to produce the reliable plasma levels required to protect against the long-term effects of oestrogen deprivation. They provide short-term relief from menopausal symptoms, in particular, atrophic vaginitis. A vaginal ring which releases estradiol at a controlled rate in physiological levels for up to 3 months is an alternative for women who cannot tolerate transdermal patches.

The dose of oestrogen used in HRT sufficient to preserve bone density is usually higher than that necessary to alleviate vasomotor symptoms. The doses suggested to protect bone density are estradiol 2 mg/day orally, 50 μg/day transdermally and 50 mg every 6 months by implant. If the conjugated equine oestrogens are used, the oral dose should be 0.625 mg/day. The lower doses found in vaginal creams may alleviate the vasomotor symptoms but will not protect against osteoporosis. Current guidelines advise the use of the lowest possible dose of HRT to relieve vasomotor symptoms and recommend alternative treatment to prevent and treat osteoporosis.

Oestrogens should be used alone only in women who have undergone a hysterectomy; if the uterus is present, the endometrium will be stimulated and this increase in endometrial growth may be a precursor to development of a malignant condition. Current practice is to administer progestogens with oestrogen. In the early 1970s, when oestrogen was used alone, HRT received a bad press because in women who had not undergone hysterectomy, there was an increased incidence of endometrial carcinoma. In women who have undergone hysterectomy, oestrogens are usually administered continuously.

Progestogen therapy

The only proven reason for adding a progestogen to oestrogen therapy for HRT in women with an intact uterus is to protect the endometrium from hyperplasia and possible neoplasia. There are many preparations which contain progestogens added to oestrogen for a number of days per month. However, to effectively prevent endometrial hyperplasia, the progestogen must be taken for a minimum of 12 days. The minimum dose of progestogen required to protect against hyperplasia depends on the potency of the compound used.

The progestogens commonly available in HRT preparations are either derivatives of progesterone, such as medroxyprogesterone and dydrogesterone, or 19-nortestosterone substitutes such as norethisterone or levonorgestrel. All these synthetic progestogens are active following oral administration and provide adequate protection of the endometrium against oestrogen stimulation. Some transdermal preparations also incorporate a progestogenic compound in the regimen. As with all semi-synthetic or synthetic hormones, these compounds may act on receptors other than the progesterone receptor, and the long-term consequence of this is not predictable.

Progesterone is the only progestogen to act solely on the progesterone receptor, but it has poor oral bioavailability and so it is difficult to achieve satisfactory plasma concentrations. However, the micronised preparations are better absorbed. Progesterone may also be administered at night in the form of a pessary or suppository, or by injection in the form of a long-lasting subdermal implant. The progestogen in HRT is most commonly administered orally or transdermally and usually one of the synthetic progestogens is used.

Oestrogen and progestogen regimens

The monthly withdrawal bleed is perceived by some post-menopausal patients to be an unacceptable side effect of HRT, and this has resulted in the development of a number of regimens in an effort to minimise this effect (Box 46.1). Formulations have been produced with which bleeding only occurs every 3 months instead of every 4 weeks, or it does not occur at all.

Box 46.1 Regimens of combined oestrogen and progestogen therapy for use in HRT

• Oestrogen 28 days + progestogen 12 or 14 days, then repeat without interval (bleed every 4 weeks)

• Oestrogen 70 days + progestogen 14 days followed by 7 days placebo tablets (bleed every 3 months)

• Oestrogen + progesterone continuously (no bleed)

• Oestrogen continuously + Mirena IUS (bleed variable but levonorgestrel likely to reduce bleed and can provoke amenorrhoea)

The use of the 70-day oestrogen preparation, while being more popular with women because bleeding only occurs every 3 months, needs further evaluation as regards endometrial protection. Bleeding can be avoided altogether if a combination of oestrogen and progestogen is given continuously throughout the treatment (continuous combined HRT). Such a preparation should only be given to women who are at least 12 months post-menopausal and have an atrophied endometrium; otherwise, breakthrough bleeding may occur. Bleeding in the first 6 months (usually just spotting) is not uncommon but bleeding after this time should be investigated, although the incidence of endometrial hyperplasia is low with this continuous regimen. Others recommend a 28-day interval between courses of treatment to allow the endometrium to become atrophic in patients who are changing from the cyclical therapy to the continuous combined therapy. Patients who commence cyclical HRT prior to ceasing menstruation should change to a continuous combined preparation only after the age of 54 (when there is an 80% chance that they will be post- rather than peri-menopausal) to reduce the risk of breakthrough bleeding. An alternative option is to use the levonorgestrol-loaded intrauterine system (IUS or Mirena) in conjunction with oral oestrogen. This option can be particularly useful for women intolerant of progestogenic side effects.

Not all progestogens have the same pharmacological profile and these differences have implications for their usage. Two of the most widely used synthetic progestogens are medroxyprogesterone acetate and norethisterone. These are used as the progestogenic component of an HRT regimen in combination with oestrogen but have been shown to increase the risk of breast cancer in long-term HRT users (Million Women Study Collaborators, 2003; Women’s Health Initiative, 2002). Structurally, medroxyprogesterone acetate is more similar to natural progesterone than norethisterone. The metabolism of these two compounds is also different, as medroxyprogesterone acetate is the major progestogenic compound rather than one of its metabolites. In contrast, the metabolites of norethisterone exhibit significant activity in addition to a wide range of non-progestogenic actions. Norethisterone also binds to sex hormone–binding globulin, whereas medroxyprogesterone acetate does not.

The most notable difference in steroid receptor-binding affinity between the two synthetic progestogens and endogenous progesterone is that although all the compounds have affinity for the mineralocorticoid receptor, only the natural compound has antagonist activity. As a consequence, the synthetic compounds may be unable to counteract the sodium-retaining and blood pressure–raising effects of the oestrogens used in HRT. Endogenous progesterone affinity for the glucocorticoid receptor is also different, with medroxyprogesterone acetate a more potent antagonist than norethisterone. This may influence their side-effect profiles and impact on inflammation, immune response, adrenal function and bone metabolism.

Tibolone

Tibolone is a synthetic steroid that has oestrogenic, progestogenic and androgenic effects that alleviate menopausal symptoms without a monthly bleed. The oestrogenic effects are weak and should not promote endometrial hyperplasia, but 10–15% of women on this treatment experience breakthrough bleeding. The drug is given continuously but is not suitable for women within 1 year of menopause or immediately after oestrogen therapy because in such cases breakthrough bleeding is most likely to occur. The evidence suggests that this drug is protective against osteoporosis, but the long-term cardioprotective effects remain unclear as there is some evidence of a lowering of HDL-C. It should also be withdrawn if signs of thromboembolic disease occur. The androgenic action of tibolone tends to increase libido. It has been reported that tibolone does have fewer breast-related adverse effects than oestrogenic or oestrogenic–progestogenic HRT regimens.

Raloxifene

Raloxifene is a non-steroidal benzothiophene that binds to some oestrogen receptors and belongs to a class of drugs referred to as selective oestrogen receptor modulators (SERMs). These compounds act selectively on some oestrogen receptors to increase bone mineral density and antagonise oestrogen-dependent effects on breast and endometrial tissues in post-menopausal women. However, there is a reported increased risk of thromboembolism, particularly in the first 4 months of use. Raloxifene cannot be used to treat vasomotor symptoms in peri-menopausal women. In fact, it is reported to induce hot flushes. It has little or no stimulatory effect on the uterine endometrium and is not associated with uterine bleeding.

In summary, raloxifene is a compound that selectively stimulates one group of oestrogen receptors and may be considered a curative treatment for osteoporosis and a preventive agent in the development of oestrogen-dependent breast tumours.

Clinical monitoring

Before initiating HRT, a detailed patient history and physical examination are essential to eliminate medical disorders and genital malignancy. Bone mineral densitometry can also be helpful to establish a baseline for subsequent measurements. Blood tests should include serum electrolytes and creatinine, liver function tests, haemoglobin, lipids and a full blood count. Urine analysis should also be performed. The history and findings from the physical examination may indicate other tests. The patient should have undergone routine cervical smear examination and, preferably, mammography. In women with an intact uterus, any irregular vaginal bleeding should be investigated to exclude endometrial pathology.

After starting therapy, the woman should be seen within 3 months in the first instance and then at intervals between 6 and 12 months so that symptoms may be assessed and any side effects of therapy can be reported. Blood pressure measurements are undertaken on these routine visits; HRT is usually associated with a fall in blood pressure due to a vasodilator action of oestrogen. Hypertension is not a contraindication to treatment with HRT but does need treatment before starting on oestrogen therapy. Some women may have an elevated blood pressure on oral oestrogen but show no such effect with the non-oral route. Weight gain may occur some months after treatment has been initiated, and the patient should be advised to reduce calorie intake accordingly.

Stopping HRT

Guidelines (see Section 7.3.1 British National Formulary) indicate that there are a number of signs and symptoms which suggest that women should be advised to immediately stop taking HRT and these include:

• sudden severe chest pain

• sudden breathlessness or cough with blood-stained sputum

• unexplained severe pain in calf of one leg

• severe stomach pain

• serious neurological effects

• hepatitis, jaundice, liver enlargement

• blood pressure above systolic 160 mmHg and diastolic 100 mmHg

• detection of a risk factor, for example, prolonged immobility after surgery or leg injury.

Once HRT has been stopped, investigation and treatment should be undertaken as appropriate.

Examples of serious neurological effects include unusual severe, prolonged headache. This is especially important if the headache occurs for the first time or it becomes progressively worse. Marked numbness, especially if it suddenly affects one side or part of the body, is important. Other neurological effects are sudden partial or complete loss of vision or disturbance of hearing or other perceptual disorders. HRT should also be stopped if the following occur: dysphasia, bad fainting attack or collapse, a first unexplained epileptic seizure/weakness and motor disturbances.

Alternatives to HRT

HRT remains the most effective treatment for vasomotor symptoms, resulting in 80–90% reduction in hot flushes. Of the non-hormonal agents, serotonin–norepinephrine reuptake inhibitors (SNRIs) such as venlafaxine and its active metabolite desvenlafaxine appear to be the next most useful treatments; gabapentin may be equally effective, whilst clonidine is only modestly effective in reducing hot flushes. Generally, there is a lack of safety information and trial data regarding alternative therapies. Trials with black cohosh, red clover and soy foods, all of which contain phytoestrogens, have yielded conflicting results and have raised concerns about hepatoxicity (black cohosh) and interactions with anti-coagulants (red clover).

Bisphosphonates such as alendronate, etidronate and risedronate are inhibitors of bone resorption and increase bone mineral density by altering osteoclast activation and function. Bisphosphonates are used with care in women with upper gastro-intestinal problems and their posology is complex. They are contraindicated in patients with hypocalcaemia. Etidronate is also contraindicated in patients with severe renal impairment.

Teriparatide, a recombinant human parathyroid hormone, is licensed to treat osteoporosis in post-menopausal women. It stimulates new formation of bone and may increase resistance to fracture. However, it has several contraindications including hypercalcaemia, severe renal impairment, metabolic bone diseases and unexplained elevations of alkaline phosphatase.

A Cochrane review supports the efficacy of strontium ranelate for the reduction of fractures in post-menopausal osteoporotic women (O’Donnell et al., 2006). The main side effect reported was diarrhoea, but the potential vascular and neurological side effects should be explored further. National guidance (NICE, 2010) details the role of alendronate, etidronate, risedronate, raloxifene and strontium ranelate for the primary prevention of osteoporotic fragility fractures in post-menopausal women.

Advances in bone biology may lead to the design of more elegant therapies to deal with osteoporosis. An example would be denosumab, a human monoclonal antibody which binds to the receptor activator of nuclear factor-κβ ligand (RANKL), which is responsible for osteoclast differentiation, activation and survival. Denosumab therefore mimics the function of osteoprotegerin limiting bone resorption.

Some small studies have shown that aerobic exercise can improve cardiovascular parameters and reduce menopausal symptoms in addition to reducing BMI.

Treatment with HRT

Vasomotor symptoms

Vasomotor symptoms include hot flushes, headaches, insomnia, giddiness and faintness. They occur in about 70–80% of women and result in physical distress in about 50%, lasting for up to 5 years in around one-quarter of the women. Flushes and sweats, particularly night sweats, indicate vasomotor instability and probably result from unoccupied oestrogen receptors on blood vessels. Oestrogens cause a rapid rise in blood flow through the blood vessels, and lack of oestrogen will render the oestrogen receptors in these vessels supersensitive to any subsequent rise in oestrogen level. During the peri-menopause and menopause, oestrogen levels tend to fluctuate and it is suggested that it is these fluctuations that result in vasomotor symptoms. The extreme sensitivity of blood vessel oestrogen receptors tends to mean that a clinical response to vasomotor symptoms is achieved with low doses of oestrogen. There is also likely to be central control via a supra-pituitary mechanism incorporating several chemical pathways involving serotonin, noradrenaline and dopamine. General advice regarding diet, for example, avoiding certain foods and drinks that cause vasodilation such as hot spicy foods and alcohol, may be helpful to some women. There is some evidence that regular exercise, which stimulates the production of hypothalamic β-endorphins, may reduce the risk of hot flushes.

Urogenital tract

These symptoms include:

• vaginal dryness and dyspareunia (painful sexual intercourse)

• vaginal discharge and bleeding

• urinary incontinence, urgency of micturition, recurrent symptoms of cystitis.

Symptoms result from oestrogen deficiency in menopausal woman and may be treated either with systemic HRT preparations or with topical applications of oestrogen incorporated into vaginal creams, pessaries and silicone vaginal rings. Such topical routes of administration do result in some systemic absorption of oestrogen through the vaginal mucosa and since this may be erratic, vasomotor symptoms may ensue. Absorption of oestrogen from these topical applications may also stimulate uterine endometrial development, if present. Consequently, it is recommended that these treatments should not be used for more than 6 months. The dose of oestrogen required to stimulate the oestrogen receptors in the vagina and the lower urethra is about 10 μcg/day, and the efficiency of such low doses has been demonstrated in a number of clinical trials. The effect of oestrogen on vaginal symptoms is more marked than its effect on urinary symptoms, but the incidence of urinary sensory dysfunction may be improved. Menopausal atrophic vaginitis may respond to a short course of a topical vaginal oestrogen preparation used for a few weeks and repeated if necessary. Current guidance is that the minimum effective dose should be used for the shortest duration.

Initial treatment with local oestrogen may result in a reaction to the presence of oestrogen in tissues containing oestrogen receptors, such as the breast and vagina. The longer the time between starting HRT and menopause, the greater these effects. During initial treatment, other effects observed include headache, appetite increase and calf muscle cramps, but these side effects usually resolve without intervention. HRT with small doses of an oestrogen (together with a progestogen in women with an intact uterus) is appropriate for alleviating menopausal symptoms such as vaginal atrophy or vasomotor instability.

Bone

In a woman not treated with HRT, approximately 15% of bone mass is lost within 10 years of menopause, resulting in an increased incidence of fracture, typically of the hip. Such fractures take up at least 10% of orthopaedic beds, and the total cost in terms of morbidity and mortality is high. The effect of oestrogen lack is to increase osteoclastic bone resorption. There is an overall increase in bone turnover, more bone is resorbed than replaced and there is an associated increase in the rate of bone loss which may continue for 5–10 years. Oestrogens may exert effects on bone through the calcium-regulating hormones such as calcitonin and parathyroid-regulating hormones. Evidence also exists for the effect of oestrogens on the local production of bone growth factors, cytokines, in particular osteoprotegerin, which blocks osteoclastogenesis and prostaglandin E₂.

The greatest effect of oestrogen on bone is seen with implants, where an approximately 8% increase in vertebral bone density is seen within 1 year of treatment. Estradiol patches are the next most effective route of administration (Fig. 46.1), while oral therapy only achieves an increase in bone density of about 2% per annum. However, 5 years of oral oestrogen therapy will still achieve a lifetime reduction in femoral neck fracture of as much as 50%.

Fig. 46.1 Percentage of change from baseline (mean ± SEM) in lumbar spine bone mineral density in patients receiving placebo (open circles) or transdermal 17β-estradio 50 μcg/day (filled circles), 75 μcg/day (squares) or 100 μcg/day (filled triangles)

(from Delmas et al., 1999).

Oestrogen improves the quality and quantity of bone in the post-menopausal woman. It may be started at any time after menopause, and the benefit will continue for the duration of treatment. Bone loss will continue after treatment ceases, which leads into the difficult area of how long treatment should continue. It was previously recommended that to obtain maximum benefit from treatment with oestrogen, the duration of the course should be at least 10 years. However, studies have identified some potential adverse effects of HRT (Million Women Study Collaborators, 2003; Women’s Health Initiative, 2002). Therefore, this recommendation is controversial, particularly in women with an intact uterus, where a combined HRT preparation is required as there is probably a greater risk of adverse outcome. Oestrogen given systemically in the peri-menopausal and post-menopausal period or tibolone given in the post-menopausal period also diminishes post-menopausal osteoporosis, but other drugs are preferred.

Raloxifene is licensed for the prevention and treatment of osteoporosis as an alternative to HRT. Raloxifene reduces bone loss and increases bone density at the spine and hip in post-menopausal women. With its oestrogen antagonist effect on breast and endometrium, raloxifene may prove to be an advance over oestrogen treatment in osteoporosis prevention and treatment in post-menopausal women. It does, however, cause hot flushes which may be unacceptable for some.

Cardiovascular system

The estimated influence of HRT on the incidence of coronary heart disease, venous thromboembolism and stroke is summarised in Table 46.3.

Table 46.3 Estimated influence of HRT on incidence of gynaecological cancers, venous thromboembolism, stroke and CHD (MHRA/CHM, 2007)

Coronary heart disease

Women at 45 years are significantly less likely than men to die of coronary heart disease, but by the age of 60 years, the death rate from the disease is similar in both sexes. Oestrogen is probably central to this gender difference as women who experience early loss of endogenous estradiol have an accelerated risk of developing coronary heart disease. Intuitively, it would appear logical that post-menopausal women would gain benefit from receiving exogenous oestrogen. However, there are conflicting views as to whether this is, or is not, the case. A 4-year study of women with coronary heart disease in the USA (Heart and Estrogen-Progestin Replacement Study, HERS) who were also receiving HRT (conjugated equine oestrogen and medroxyprogesterone acetate) demonstrated no significant reduction in stroke (Simon et al., 2001). This was in contrast to the increased risk of myocardial infarction in the first year of starting HRT in the original HERS report (Hulley et al., 1998). The Women’s Health Initiative (2002), another US-based study, also reported an increase in coronary events in the first year of receiving HRT.

The relevance of both studies to UK practice has been challenged as both involved use of conjugated equine oestrogen/medroxyprogesterone acetate regimens. In the UK, natural progesterone is preferred, and this differs from medroxyprogesterone acetate in that it does not attenuate the beneficial effects of oestrogen in reducing the development of coronary artery atherosclerosis and protecting against coronary vasospasm. A study (De Vries et al., 2006) using the UK General Practice Research Database found HRT was associated with a decrease in risk of acute myocardial infarction, and there was no difference between the different oestrogen–progestogen combinations. A recent study found that transdermal but not oral oestrogen replacement therapy significantly reduced atherogenic index of plasma (Vrablik et al., 2008). This occurred by increasing HDL particle size and therefore improving the antiatherogenic properties.

The International Menopause Society consensus statement (2009) states that HRT can be given to women around the age of natural menopause without increasing the risk of coronary heart disease and may even decrease the risk in this age group. HRT is not contraindicated in women with hypertension, and in some cases, HRT may even reduce blood pressure. HRT is contraindicated in women with a history of myocardial infarction, stroke or pulmonary embolism.

Venous thromboembolism

It is known that the oestrogen in the combined oral contraceptive pill contributes to thromboembolic disorders. Whether the same applies to the oestrogen in HRT is unclear as the dose of oestrogen found in HRT is much lower and more physiological than that in the oral contraceptive pill. Analysis of the results from the Women’s Health Initiative (2002) has confirmed that there is an increased risk of pulmonary embolism with HRT and the risk is greater with oral HRT than transdermal oestrogen.

Oral oestrogen, in contrast to the transdermal route, undergoes extensive first-pass metabolism. This increases the production of prothrombotic factors in the liver and is associated with a reduction in fibrinogen and factor VII activation, such as von Willebrand’s factor and anti-thrombin, and enhanced fibrinolysis (RCOG, 2004). HRT is also associated with increased resistance to activated protein C.

Whilst oestrogen exposure and age increase the risk of venous thromboembolism, they do not explain the increased risk seen in the first year of HRT use. An underlying thrombophilia, a known risk in hyperoestrogenic situations, is thought to be the additional factor that makes certain women susceptible to venous thromboembolism. Guidelines (RCOG, 2004) to manage the risk of venous thromboembolism have been produced for women starting or continuing HRT (Box 46.2). The selective oestrogen receptor modulators (SERMS) such as raloxifene are considered to carry the same risk of thrombosis as oestrogen-containing HRT.

Box 46.2 Summary of guidelines for dealing with the risk of venous thromboembolism (VTE) in women receiving HRT

(adapted from RCOG, 2004)

Women starting or continuing HRT

• Women starting or continuing HRT should be counselled with regard to perceived benefits and possible risks for their individual situations, including consideration of alternative therapies.

• All women commencing HRT should be counselled about the risk of VTE, should be aware of the signs and symptoms of VTE and should be able to access medical help rapidly if they suspect that they have developed a thrombus.

• Universal screening of women for thrombophilic defects prior to or continuing the prescription of HRT is inappropriate.

• Prior to commencing HRT, a personal history and a family history assessing the presence of VTE in a first- or second-degree relative should be obtained.

• HRT should be avoided in women with multiple pre-existing risk factors for VTE.

Women with a personal or family history of VTE

• Testing for thrombophilia should be discussed with, and available for, women with a personal or family history of VTE.

• In women with a previous VTE, with or without underlying heritable thrombophilia, oral HRT should be usually avoided in view of the relatively high risk of recurrent VTE.

• In women with a personal history of VTE but with an underlying thrombophilic trait identified through screening, HRT is not recommended in high-risk situations such as anti-thrombin deficiency or with combinations of defects.

• In women over 50 years with a history of VTE within the previous year, a full clinical history and examination with appropriate investigations is warranted for underlying disease.

• If a woman who is on HRT develops VTE, HRT should be discontinued.

• If a woman wants to continue on HRT after VTE, long-term anti-coagulation should be considered.

Overall, minimising cardiovascular risk whilst obtaining the benefits of HRT is influenced by the age of the women treated, their body mass index, cardiovascular health and menopausal history, the timing of initiation of HRT, the formulation of the product used, and polymorphism.

Cancer

Details of the estimated influence of HRT on the incidence of cancer are presented in Table 46.3.

Colorectal cancer

Colorectal cancers tend to occur in women over 50, and several studies have indicated that HRT can reduce the risk by around 35%. The exact mechanism for this protective effect on the colon is unclear, although it has been suggested that oestrogen may decrease the formation of potentially carcinogenic bile acids.

Ovarian cancer

The risk of ovarian cancer risk is known to be affected by exogenous hormones, with a reduced incidence consistently reported in users of oral contraceptives. The relationship between the hormones used in HRT and ovarian cancer is, however, less clear. Women who use oestrogen plus a progestogen do not appear to be at increased risk of ovarian cancer, perhaps because the progestogenic component has an apoptotic effect on the ovarian epithelium. This has been supported by studies in small groups of women with a mutation in either the BRCA1 or BRCA2 gene (point mutations associated with breast and ovarian cancer) which demonstrated that HRT did not adversely influence the risk of developing ovarian cancer. In contrast, other large HRT trials have suggested a possible increase in risk of ovarian cancer, particularly in women who have had a hysterectomy and are receiving long-term oestrogen-only HRT.

Endometrial cancer

Unopposed oestrogen therapy at least doubles the risk of endometrial cancer, but the addition of a progestogen, either in continuous combined or sequential regimens, reduces this risk. Of the two regimens, continuous combined administration is more effective than sequential therapy in reducing the risk of endometrial hyperplasia.

Breast cancer

Breast cancer is the most common cause of disease and death in middle-aged women, affecting around one in 11 women before the age of 75 in developed countries. Over the past 20 years, there has been a growing body of evidence to suggest that progesterone contributes to the development of breast cancer. Evidence to support this came from trials in which use of combined regimens increased breast cancer risk more than oestrogen alone. This is thought to have occurred because the synthetic progestogens possessed some non-progesterone like effects that potentiated the proliferating action of oestrogen. In the USA, the most commonly used progestogen in HRT was medroxyprogesterone acetate combined with conjugated equine oestrogen and administered orally. In contrast, in central and southern Europe, a wider range of progestogens is used, particularly the 19-nortestosterone derivatives. Studies investigating the use of natural micronised progesterone to improve bioavailability in HRT regimens have demonstrated no increased risk of breast cancer. In vitro studies with medroxyprogesterone have found it promotes the reproductive transformation of estrone into estradiol by influencing the activity of 17β-hydroxysteroid dehydrogenase. The properties of the synthetic progestogens are outlined in Table 46.4.

Table 46.4 Properties of progestogens and their link to breast cancer

Progestogen	Action
19-Nortestosterone derivatives, medroxyprogesterone acetate	Oestrogenic activity Influence on 17β-hydroxysteroid dehydrogenase
19-Nortestosterone derivatives, medroxyprogesterone acetate	Metabolic effects, opposing those of oestrogen, on insulin sensitivity. Hepatic effects, opposing those of oestrogen, that is increasing IGF-1 (insulin-like growth factor), decrease in sex hormone-binding globulin
19-Nortestosterone derivatives	Binding to sex hormone-binding globulin with consequent reduction in capacity to bind oestrogen

(adapted from Campagnoli et al., 2005)

The relationship between different HRT regimens and breast cancer is complex (Table 46.4). There are many potential confounders to be considered when interpreting trial data, for example, time of initiation of therapy, in relation to menopausal status, body mass index, prior hormone use, etc. Chlebowski et al. (2009) reported on the decline in breast cancer in the USA after a reduction in HRT usage following publication of the WHI trial (2002). This prompted a response from the President of the International Menopause Society (Sturdee, 2009) summarised as follows:

• Decline in breast cancer rates started at least 3 years before the WHI study was halted (which led to a dramatic drop in HRT use). This implies that decline in breast cancer rates must be independent of the reduction in HRT use.

• Breast cancer takes years to develop and at least a decade to reach the detectable stage. If HRT use causes breast cancer, then the drop in breast cancer rates would not be seen for some time.

• The drop in breast cancer rates could be due to other factors, for example, changes in screening.

Notwithstanding these arguments, the problem on what to advise women contemplating HRT can be problematic. Current advice recommends that the minimum effective dose of HRT should be used for the shortest duration, but for the treatment of menopausal symptoms the benefits of short-term HRT outweigh the risks in the majority of women, especially in those aged under 60 years (Joint Formulary Committee, 2010).

Psychological symptoms

Psychological symptoms include:

• depression

• mood changes and irritability, which may also be associated with other life changes occurring at this time

• exhaustion

• poor concentration and memory

• panic attacks

• lowered libido, which may be exacerbated by the above symptoms together with dyspareunia linked to lack of oestrogen and falling androgen levels.

The role of HRT in this area has not been clearly defined, although in several studies, surgical menopause has been associated with depression, indicating a correlation with oestrogen lack. Many women experience psychological symptoms around menopause, and although these may be associated with oestrogen deficiency, they may also result from the changes in family life that often occur around this time. Disturbance of sleep pattern and sleep deprivation, associated with menopause, are likely to contribute to the psychological symptoms. Many women find treatment with estradiol restores normal sleep and psychological problems are then reduced. Some of the mood changes will respond to counselling and psychotropic drugs. Treatment with oestrogens at high doses (patches of 100 μcg or implants of 50 mg) has also been shown to improve depression scores. If a progestogen is added into the regimen, then the results are less predictable since progestogen use is related to mood changes, particularly in women who have previously suffered from the pre-menstrual syndrome. Age negatively influences almost all sexual function domains in a significant manner.

HRT improves some aspects of sexual function during menopause, but it does not appear to improve the domains of desire and arousal. The lowered libido experienced during menopause is associated with reduced levels of circulating androgen resulting from ovarian failure. It has been demonstrated that subcutaneous implants of testosterone, 100 mg every 6 months, will increase the libido in a high proportion of patients.

Central nervous system

The relationship between oestrogen and neurodegenerative conditions, in particular, Alzheimer’s disease (Mulnard et al., 2000), has received attention in the light of an observation that there is an increased incidence of the disease in older women. The development of plaques of amyloid-β, a protein that disrupts nerve cell connections in the brain, occurs more rapidly in the absence of oestrogen. This effect of amyloid-β production results in the symptoms of short-term memory loss and disorientation which occur in Alzheimer’s disease (AD). Further studies are essential to clarify the relationship between HRT and Alzheimer’s disease. There is some evidence that heavier women, who have higher free estradiol levels, exhibit better cognitive function in several domains. Such women may also have a greater clinical response when using exogenous oestrogen. It should also be noted that conjugated equine oestrogen is composed primarily of oestrone sulphate and this oestrogen has a much lower affinity for oestrogen receptors than oestradiol. Hence, the findings with respect to exogenous oestrogen and neurological function remain equivocal. The length of oestrogen deprivation before supplementation is also important as early administration of oestrogen for a period of several years may yet be found to be beneficial. Evidence is emerging that the progestogenic component in combined HRT is important and medroxyprogesterone acetate usage has been associated with negative cognitive outcomes. The potential of selective oestrogen receptor modulators (SERMs) as neuroprotectants has been evaluated in a breast cancer prevention trial which compared tamoxifen with raloxifene (Yaffe et al., 2001). Both agents were found to have similar effects on cognition. It is possible that HRT may have a neuroprotective effect under certain circumstances in some women and neuroimaging, for example, PIB (Pittsburgh compound B – a fluorescent agent). PET scanning may reveal effects not detectable using cognitive testing.

Case studies