A local audit evaluating bone health in patients with functional hypothalamic amenorrhoea secondary to an eating disorder and a review of the application of hormone therapy in this clinical setting

Abstract

It is widely known that estrogen has a fundamental role to play in skeletal homeostasis. In the most reductionist sense, the action of estrogen can be surmised as anti-resorptive. Estrogen prevents the break-down of bone. It therefore follows that estrogen deficiency states, such as the menopause and functional hypothalamic amenorrhoea (FHA), are often characterised by increased bone remodelling and disrupted skeletal homeostasis. FHA is the cessation of menstruation secondary to abnormal signalling between the hypothalamus and pituitary gland due to deficient pulsatile secretion of Gonadotrophin Releasing Hormone (GnRH). Functional hypothalamic amenorrhoea is frequently a consequence of women suffering with eating disorders. The development of FHA secondary to eating disorders is an evolutionary adaptive response to chronic metabolic energy deficiency. Fundamentally, preservation of life is biologically prioritised over dispensable physiological process such as reproduction. Consequently, the hypothalamic-pituitary-ovarian (HPO) axis fails, which disrupts menstrual function and ovulation, culminating in a state of estrogen deficiency. One of the most important and long-lasting deleterious consequences of FHA is disrupted skeletal homeostasis and bone loss. Estrogen replacement, most commonly in the form of combined hormone replacement therapy (HRT) or the combined oral contraceptive pill (COCP), is advised for women with an early menopause to prevent bone loss. Arguably, estrogen replacement should also be utilised in the context of FHA. However, the optimum estrogen regime for women with FHA remains under-researched and so management is not evidence-based.

Keywords

Functional hypothalamic amenorrhoea Menopause Osteoporosis Eating disorder Hormone replacement therapy

Introduction

Estrogen has a fundamental role to play in skeletal homeostasis.^1–5 In the most reductionist sense, the action of estrogen can be surmised as anti-resorptive. In other words, estrogen prevents the break-down of bone. It therefore follows that estrogen deficiency states, such as the menopause and functional hypothalamic amenorrhoea (FHA), are often characterised by increased bone remodelling and disrupted skeletal homeostasis. Indeed, bone mineral density (BMD) decreases by approximately 2% per year in the early perimenopause phase, associated with falling estrogen.⁶ FHA is the cessation of menstruation secondary to abnormal signalling between the hypothalamus and pituitary gland due to deficient pulsatile secretion of Gonadotrophin Releasing Hormone (GnRH). The reduced secretion of GnRH leads to a reduced secretion of Follicle Stimulating Hormone (FSH) and Luteinising Hormone (LH), culminating in a failure of normal ovarian function and estrogen deficiency.¹ Parallels can be made between an early menopause, which is known to have potentially devastating effects on long-term bone health, and FHA. Estrogen replacement, most commonly in the form of combined hormone replacement therapy (HRT) or the combined oral contraceptive pill (COCP), is advised for women with an early menopause to prevent bone loss.^4,5 Arguably, estrogen replacement should also be utilised in the context of FHA. However, the optimum estrogen regime for women with FHA remains under-researched and so management is not evidence-based.

Functional hypothalamic amenorrhoea is frequently a devastating consequence of women suffering with eating disorders. Eating disorders, such as anorexia nervosa (AN), place stress on the hypothalamic-pituitary-adrenal (HPA) axis and the autonomic nervous system. This results in a constellation of neuroendocrine and metabolic alterations, including hypothalamic hypothyroidism, which aim to conserve and divert energy expenditure.² The development of FHA secondary to eating disorders can be thought of as an evolutionary adaptive response to chronic metabolic energy deficiency. When energy is scarce, the physiological mechanisms that partition energy will favour those processes that ensure survival of the individual over processes such as reproduction.^2,3 Consequently, the hypothalamic-pituitary-ovarian (HPO) axis fails, which disrupts menstrual function and ovulation, culminating in a state of estrogen deficiency and bone loss.

Functional hypothalamic amenorrhoea encompasses an extensive number of hormonal and metabolic changes, the full description of which goes beyond the scope of this audit. Perturbations to several hormonal pathways are seen including the HPA axis, HPO axis and the hypothalamic-pituitary-thyroid axis. A stressed HPA axis is associated with hypercortisolaemia. Hypercortisolaemia is directly deleterious on the bone due to inhibition of osteoblast proliferation and bone formation.⁶ Blunted FSH secretion during the luteal-follicular transition may predispose to luteal defects and progesterone deficiency.² Progesterone stimulates bone formation; therefore its deficiency will be detrimental to the accretion of peak bone mass and skeletal homeostasis.⁷ Additionally, patients with eating disorders commonly have low amounts of adipose tissue. A minimum amount of adipose tissue is needed for preservation of reproductive function, and therefore any eating disorder that compromises this will impact amenorrhoeic status.²

The lifecycle of BMD is largely under the influence of the HPO axis and therefore changes throughout a woman’s reproductive lifespan. Estrogen is a potent stimulator of bone mineral accretion through puberty.⁵ In children and adolescents, BMD steadily increases through bone modelling-related gains. Peak bone mass is acquired during the menstruating years with a peak total hip and femoral neck BMD acquired during the ages 16–19 years.⁷ During the pre-menopausal years, BMD remains steady through a state of persistent bone formation and breakdown. As women enter the perimenopause, estrogen levels decline and bone resorption starts to overbalance bone formation and life’s most rapid bone loss is seen. Typically eating disorders affect young pre-menopausal women and therefore, by inducing a hypo-estrogenic state, will directly interfere with peak bone mass accretion during adolescence and early adulthood. Failure to achieve peak bone mass during these years increases the risk of osteoporosis later in life. Furthermore, low BMI correlates with lower BMD.⁶ Therefore, one of the most important and often long-lasting deleterious consequences of FHA is poor bone health. Therefore, these women represent a cohort of patients who are high risk for future bone disease and fractures later in life, and in whom estrogen replacement needs to be considered. It has been reported that young women with eating disorders are known to be at a sevenfold higher risk of fracture.² 57% of women with AN will sustain at least one fracture in their lifetime.⁶ Between 10 and 20% of people who sustain hip fractures die within 1 year, and increased mortality is also seen after vertebral and other non-vertebral fractures.⁸

Conceivably, women with FHA are equally at risk of osteoporosis compared with women with premature ovarian insufficiency (POI). POI is loss of ovarian function under the age of 40 years. It is well known that women with POI are advised to have estrogen treatment, most commonly in the form of HRT or the COCP, to prevent bone loss.⁴ However, estrogen replacement is not commonly considered for women with FHA. There is no agreed consensus on ideal hormone replacement regimes for women with POI and even less available evidence to guide practice in FHA.

In clinical practice, women with FHA are often prescribed estrogen replacement in the form of the COCP. The COCP most commonly contains synthetic ethinylestradiol (EE), in combination with a synthetic progestin. In contrast, HRT most commonly contains 17β-estradiol, which is found naturally and therefore, considered physiological. These different estrogen formulations may have differing benefits and risks. Furthermore, 17β-estradiol can be delivered orally or transdermally, which confers further pharmacokinetic differences. Oral 17β-estradiol is absorbed from the gastrointestinal tract and passes first through the portal circulation. This first-pass through the liver can lead to 60–90% of the drug being metabolised, mainly to estrone and its conjugates. Transdermal HRT avoids first past metabolism in the liver and the subsequent conversion to estrone.⁹ These important pharmacokinetic differences between different estrogen formulations may explain the differing effects on bone. EE delivers a pharmacological rather than physiological replacement estrogen dose. The COCP modulates endogenous hormone production and suppresses ovulation. Therefore, it is a useful treatment option for this younger cohort of women who need bone protection and contraception. However, the provision of physiological estrogen replacement in the form of HRT containing 17β-estradiol and a cyclical progesterone may be a superior option for girls with FHA owing to the fact several randomised control trials (RCT) comparing HRT and the COCP in women with FHA or POI indicate superiority of HRT with respect to improving BMD.^2,4–6 The European Society for Endocrinology advocate the use of HRT in preference to the COCP for the purpose of bone health in women with FHA secondary to eating disorders.² The European Society of Human Reproduction and Embryology (ESHRE) acknowledges that physiological estradiol replacement may have a greater beneficial effect on BMD compared to synthetic estrogen found in the COCP for women with POI.¹⁰ However, it is important to recognise that whilst parallels can be drawn between POI and FHA, owing to the fact both represent young cohorts of women with estrogen deficiency, the prevention of bone loss in FHA is likely to be more challenging owing to the greater complexities of this endocrine disorder (e.g. hypercortisolaemia and hypothyroidism) and the implications of malnutrition and/or excessive exercise on bone health. In women with FHA, The National Institute for Clinical Excellence (NICE) guidelines recommend clinicians ‘consider offering hormone replacement therapy if amenorrhoea persists for more than 12 months’.¹¹ However, no gold-standard is referred to. There is a lack of robust evidence to help guide women to make an informed decision about appropriate estrogen replacement and more research is needed.

Bisphosphonates are commonly used for the prevention and treatment of osteoporosis. This class of drug effectively reduces bone resorption, increases BMD, and reduces risk of vertebral fracture by 50–60% and non-vertebral fracture by 20–25% in women with osteoporosis.¹² However, their use is limited in FHA for several reasons. Firstly, bisphosphonates are synthetic analogues of pyrophosphates, skeletal building blocks, which allow them to be incorporated into the skeleton. This means they have a very long half-life and the risks associated with skeletal retention are not well elucidated.^12,13 As such, their use in women with FHA secondary to eating disorders, who typically represent a younger cohort of women, has not been widely evaluated. Another barrier to the use of bisphosphonates in FHA is a small teratogenic risk and neonatal complication risk in a population often in their reproductive years.^2,14 No positive effects of bisphosphonates have been observed in adolescents with anorexia nervosa.¹⁴

In this retrospective audit analysis of the data registry from a general practice of 22,000 patients, I identified that only 17% of women suffering with eating disorders were asked about their periods at the time of diagnosis, and bone health was discussed in only 10% of consultations. Nearly one third of patients with an eating disorder developed secondary FHA and only 50% of these patients underwent a dual-energy X-ray absorptiometry (DEXA) scan. All patients who received a DEXA scan were diagnosed with osteopenia and/or osteoporosis, which highlights the importance of addressing bone health in these women. No patient received physiological estradiol replacement as the first treatment option.

Aim and objectives

The aim of this audit was to evaluate how many patients suffering with FHA, secondary to an eating disorder, were offered a DEXA scan to evaluate their BMD and estimate their future fracture risk, and to assess the treatment options offered in those patients found to have osteopenia and/or osteoporosis.

The objectives of this audit were:

1. To identify the number of patients diagnosed with FHA secondary to an eating disorder

2. To identify the proportion of patients with an eating disorder who underwent a DEXA scan to evaluate their bone health

3. To analyse the treatment options offered to those patients diagnosed with osteopenia and/or osteoporosis

Standards

• The Endocrine Society suggest that a baseline BMD measurement by DEXA scanning is performed for any adolescent or woman with 6 or more months of amenorrhoea, and that clinicians obtain it earlier in those patients with a history or suspicion of severe nutritional deficiency, other energy deficit states, and/or skeletal fragility.²

• NICE CKS states that a DEXA scan should be offered in those under 40 years of age who have a major risk of fragility fracture. A fragility fracture risk score, such as those provided by the FRAX online assessment calculator, is not needed.¹¹

• The National Osteoporosis Foundation and the International Society of Clinical Densitometry both recommend a DEXA for women under the age of 65 years who have clinical risk factors such as low body weight or a disease or condition associated with bone loss.¹⁵

• NICE CKS recommend clinicians consider using HRT for women with FHA if amenorrhoea persists for more than 12 months to manage the risk of osteoporosis associated with this physiological state of estrogen deficiency.¹¹

• The Endocrine Society advises against patients with FHA using oral contraceptive pills for the sole purpose of improving BMD. The Endocrine Society suggests clinicians consider HRT, in the form of transdermal estradiol therapy with cyclic oral progestin, for women with FHA whose menses have not resumed despite 6–12 months of nutritional, psychological, and exercise-related interventions.²

Method

A retrospective report was run on 22,000 patients registered with a NHS GP Surgery in Leigh-on-Sea, Essex. The report identified all patients that had been diagnosed with an eating disorder. It is important to highlight that the diagnosis was made by the general practitioner and not an eating disorder specialist. Eating disorders are a broad group of psychological disorders associated with abnormal eating behaviours. In this audit, the diagnosis ‘eating disorder’ referred to anorexia nervosa (AN), bulimia nervosa (BN), or binge eating disorder (BED). 29 patients were identified with a diagnosis of eating disorder. 86% of patients had AN, 10% had BED, and 4% had a diagnosis of BN. The notes of each patient were reviewed individually to identify:

• those patients who had discussed their periods at the time of diagnosis, and which patients were subsequently diagnosed with FHA

• those patients who had a discussion of bone health during any consultation post-diagnosis of an eating disorder

• those patients who were offered a DEXA scan

• the treatment offered to those patients found to have osteopenia and/or osteoporosis

Results

This retrospective audit identified 29 patients with a diagnosis of an eating disorder. The majority of patients (86%, n = 25) had a diagnosis of AN. The average age at diagnosis was 25 years and the median age was 21 years. On average, the length of time passed since diagnosis was 8 years. However, the longest period of time since diagnosis was 54 years and the shortest period of time was 1 year. The median number of years since diagnosis was 4 years. The mode was 4 years. Interestingly, this coincides with 2020, a year of national lockdown due to the SARS-CoV-2 virus. Our small data set corroborates the findings from a number of studies that have consistently demonstrated that the pandemic lockdowns were associated with worsening of eating disorders.¹⁶

Proceeding a diagnosis of one of the three aforementioned eating disorders, almost one third of patients went on to develop FHA (see Table 1 for a breakdown of the results). In other words, 27% of women with an eating disorder encountered complete cessation of their normal menses. It is likely that this number under-represents the true number of women with FHA because 14% of patients diagnosed with an eating disorder were concomitantly taking the COCP, which may have led to a false misinterpretation of a pill-induced withdrawal bleed as normal menses. The combined contraceptives used were: Microgynon (ethinylestradiol 30 mcg, levonorgestrel 150 mcg) and Gedarel (ethinylestradiol 20 mcg, desogestrel 150 mcg).

Table 1.

Summary of results.

Summary of results	n	%
Total number of patients with an eating disorder	29	100
Patients who developed FHA	8	27.6
Patients with FHA who underwent a DEXA scan	4	13.8
Patients without FHA who underwent a DEXA scan	0	0.0
DEXA scan results (n = 4)
Normal BMD	0	0
Osteopenia	1	25
Osteoporosis	3	75
Treatment given for those with osteopenia and/or osteoporosis (n = 4)
Bisphosphonates	1	25
COCP	1	25
Calci-D	2	50
HRT	0	0
No treatment	0	0
Patients on the COCP at the time of ED diagnosis:	4	13.8
Number of patients asked about their periods at time of ED diagnosis:	5	17.2
Number of consultations in which bone health was discussed and documented at any time point:	3	10.3

Despite the well-acknowledged link between FHA and osteoporosis, only half of patients with FHA underwent a DEXA scan. All patients that underwent a DEXA scan were found to have a lower-than expected BMD, meeting the criteria for a diagnosis of osteopenia or osteoporosis. 75% of patients were found to have osteoporosis and 25% of patients were diagnosed with osteopenia. HRT was not the first-line treatment for any of these patients. The treatments given to those diagnosed with osteoporosis included bisphosphonates, the COCP (containing EE) and calcium and vitamin D supplementation (calcium carbonate and cholecalciferol).

A discussion surrounding bone health was documented in only 10% of consultations. At the time of diagnosis, only 17% of patients were asked about their periods.

Discussion and conclusion

The achievement of peak bone mass is important to bone health and plays a vital role in preventing osteoporosis and subsequent fractures years later. Hip fractures could be reduced by 30% if peak bone mass was increased by 10%. The sex hormones, body weight, and nutrition play a pivotal role in peak bone mass accumulation, which explains why FHA secondary to an eating disorder can have significantly deleterious effects on bone health.^6,17 In women suffering with AN, estrogen deficiency is thought to be the primary contributing factor to bone loss.⁶ This explains why all patients that underwent a DEXA assessment were found to have lower than expected BMD.

Estrogen in the broadest sense can be thought of as anti-resorptive. However, the relationship between estrogen and bone is complex and the role of estrogen goes far beyond direct induction of osteoclastic apoptosis and inhibition of resorption. Estrogen accelerates calcium resorption by the intestine and reduces calcium excretion from the kidney with the net effect of increasing total body calcium. Estrogen also stimulates the production of calcitonin and lowers the sensitivity of bone to parathyroid hormone, both of which further inhibit bone resorption. Finally, estrogen exerts an anti-apoptotic effect on osteoblasts.¹⁷ In hypo-estrogenic states, there is diminished inhibition of osteoclastogenesis and osteoclast activity allowing the balance of skeletal remodelling to be tipped in favour of bone resorption.¹⁸

Acknowledging the fundamental relationship between estrogen and bone, estrogen replacement therapy appears to be a logical treatment modality for the preservation of BMD and reduction of future fragility fracture. However, there is a lack of robust data to help guide decision-making for women. The studies that are available suggest that HRT may have more favourable effects on bone density than the COCP.^2,4,6,7,10 Prospective clinical trials using EE-containing COCP in women with AN did not have any effect on BMD after 1 or 2 years of therapy. Conversely, use of 100 μg transdermal 17β-estradiol has been shown to lead to near-normalisation of BMD after 18 months of therapy.^2,6 A recent study randomised girls aged 14–25 years with FHA to receive either 100 mcg 17 β-estradiol patches, a COCP containing 30 μg EE or no estrogen. Only those who received the transdermal patches exhibited improvements in BMD at 12 months.¹⁴ Similar trends have been demonstrated in studies comparing the effectiveness of HRT and COCP in women with POI. An open-label RCT comparing HRT and the COCP on bone density in spontaneous POI demonstrated the use of HRT to be associated with a significant increase in BMD, the use of COCP was associated with no change in BMD and the no treatment group experienced a decrease in BMD.⁴ This significant BMD improvement was replicated in a retrospective review of 20 women with non-genetic POI, taking oral HRT over 3 years, whilst the women using transdermal HRT showed a non-significant increase in BMD and those using the COCP saw a decrease.¹⁹ It is important to highlight that either treatment is clearly superior to no estrogen replacement at all which is associated with bone loss.^4,20 Furthermore, an exploratory study comparing HRT and the COCP in young women with POI demonstrated a significant improvement in lumbar spine BMD after 12 months of HRT but not following 12 months of treatment with COCP.⁵ More research is needed to assess the lifetime risk of fracture in women with FHA and to compare the relative effectiveness of HRT (any route) compared to the COCP with respect to BMD. The Premature Ovarian Insufficiency Study of Effectiveness of hormone therapy (POISE) trial is a multi-centre prospective RCT designed to determine if HRT is superior to the COCP in relation to BMD in women with POI, based on the hypothesis that HRT provides more physiological continuous hormone supplementation with natural estrogen.¹⁹ The POISE trial is a much-anticipated RCT which will shed light on which estrogen replacement regime is most effective at ensuring future bone health. Whilst the study is not assessing women with FHA, the results can be extrapolated and will undoubtedly help clinicians and patients make more well-informed decisions about estrogen replacement in hypo-estrogenic disease states.

There may be several plausible reasons for the inferiority of the COCP compared with HRT. Firstly, if the COCP is taken in the conventional manner, there is a 7-day break from estrogen every 28 days, whereas with HRT, the estrogen is continuous. It is also possible synthetic EE has different tissue-specific effects compared with 17β-estradiol. The differing pharmacokinetics between different estrogen replacements, as described earlier, will play a role. The COCP inhibits IGF-1 production via first-pass hepatic metabolism, from which transdermal estradiol is exempt. IGF-1 is a bone-trophic, nutrition-dependent factor that stimulates osteoblastic function and bone formation.^2,6 Reduced IGF-1 levels are associated with a fall in procollagen type 1 N-terminal propetide (P1NP) levels, which is a biomarker for osteoblastic activity. Finally, oral estrogens, including both synthetic EE and 17β-estradiol, increase sex hormone binding globulin (SHBG), thereby reducing bioavailable estrogen to the detriment of skeletal homeostasis.^2,14 However, oral 17 β-estradiol has less impact on the synthesis of hepatic proteins compared with EE and so may exert less effect on IGF-1 and SHBG synthesis.²¹

In this audit, 4 patients were taking the COCP at the time of being diagnosed with an eating disorder. These patients may be falsely reassured by a monthly hormone-induced withdrawal bleed which may mask amenorrhoea. One patient diagnosed with osteoporosis was treated with Dianette, synthetic EE and cyproterone acetate. Whilst HRT may be a better treatment option for long-term bone health, the COCP may be more socially acceptable than HRT to young women with FHA. The COCP, unlike HRT, also provides contraception and is free of prescription charges in England. Ultimately, the individual should be allowed to make a well-informed decision about which treatment is best for them, but to do this more research is needed. Importantly, it is unknown if the bone-preserving effects of HRT will be observed if disordered eating patterns continue,² which highlights the importance of specialist multi-disciplinary care for women suffering with eating disorders. The treatment of eating disorders is complex and will include a comprehensive biopsychosocial approach which goes beyond the scope of this audit.

The fundamental role of concomitant progesterone is to provide endometrial protection and thereby prevent endometrial hyperplasia under the influence of unopposed estrogen therapy. However, progesterone appears to play a physiological role alongside estrogen in achieving peak bone mass. Several studies have shown the ability of progesterone to promote osteoblast differentiation, maturation, and numbers. A 1 year randomised, double-blind, placebo-controlled trial demonstrated that cyclic administration of the synthetic progestin medroxyprogesterone acetate (MPA) 10 mg/day, increases BMD in pre-menopausal women with FHA. Therefore, it is plausible that the co-administration of cyclic progesterone with transdermal estradiol may have therapeutic benefits beyond the role of endometrial protection.²² The optimal regimen for progesterone replacement is unknown. The progesterone component is important to consider and may also offer an explanation regarding the superiority of HRT compared with COCP in bone preservation. The COCP contains a higher progestin dose than sequential HRT. Higher progestin doses may adversely affect bone by binding to glucocorticoid receptors and inhibiting osteoblasts.⁴

Two patients were given treatment with Calci-D after being diagnosed with osteopenia and osteoporosis, respectively. Whilst calcium is fundamental for the development of peak bone mass in children and adolescents and should be considered for all women nutritionally deplete secondary to their eating disorder, there is little evidence to suggest that calcium and vitamin D treatment alone will treat bone loss. A large meta-analysis published in JAMA looked at 33 randomised trials involving 51,145 people comparing the incidence of fractures in people taking supplements versus those taking a placebo or getting no treatment. This meta-analysis showed no association between calcium and vitamin D supplementation and the incidence of fractures.²³ In the context of FHA secondary to an eating disorder, Calci-D is likely to represent a suitable adjunct in therapy for osteopenia and/or osteoporosis but will not provide sufficient protection from future fragility fractures in isolation. Both estrogen and progesterone stimulate bone formation and inhibit bone loss and, to date, no therapy studied has been better than HRT in preventing osteoporosis and fractures in the spine and hips.²⁴ Therefore, it seems plausible and sensible that women with FHA should be offered HRT to help future proof their bone health.

One patient was diagnosed with an eating disorder at age 52. She continued to have periods at this time. One year post-diagnosis a DEXA scan was performed which identified osteopenia. No treatment was given. The DEXA scan was repeated after 4 years which showed continued bone loss reaching the threshold for diagnosis of osteoporosis. This demonstrates that any estrogen treatment is always superior to no treatment at all. At age 57, the patient was treated with a bisphosphonate. Whilst the limitations of bisphosphonates pertaining to teratogenic and neonatal risk are less relevant in this case, the extensive number of side effects of bisphosphonates remains true. There are risks of over-suppression of bone remodelling leading to femoral fragility fractures and development of osteonecrosis of the jaw. Therefore, this class of drugs should be avoided in younger patients under 65 years of age.^12,13 Furthermore, there is no clinical evidence that attests the efficacy or safety of these treatments in young women with FHA and eating disorders, or in the context of POI.¹² Arguably, this patient would have been better treated with HRT, which would not only preserve bone mass and reduce the risk of future fracture but would also confer benefit on the cardiovascular system and reduce the risk of colorectal cancer. Furthermore, Cochrane analysis suggests that systemic HRT started before the age of 60 years or within 10 years of the menopause is associated with a reduction in all-cause mortality.²⁵ For these reasons, women with POI are actively encouraged to use HRT at least until the average age of the menopause. It stands to reason, that women with low estrogen states, such as FHA, may also be encouraged to take HRT for these reasons.

However, the decision to start HRT must follow a comprehensive discussion considering individualised benefits and risks. Venous thromboembolism and breast cancer have been associated with some estrogen replacement regimes. From a theoretical perspective, it is unlikely that the use of physiological dose of estradiol in women with FHA leads to an increased risk in breast cancer compared with women with normal ovarian estrogen production. This is supported by the fact that the use of HRT in women with POI is not associated with changes in breast density, a marker of increased breast cancer risk.¹⁰ The risk of venous thromboembolic disease with oral estrogen, including both estradiol and synthetic estrogens, can be avoided with the use of transdermal estradiol.²⁶ The risk of VTE in a minority sub-group of women with POI from the Women’s Health Initiative study was not increased.¹⁰ Therefore, it is unlikely to be a concern in young women with FHA without other predisposing factors to thromboembolic disease.

Bone health was discussed with only a small proportion of women diagnosed with an eating disorder in this audit, regardless of their menstrual status. Both eumenorrhoeic and amenorrhoeic women with eating disorders display lower than normal BMD.² Clinical tools, such as FRAX and QFracture, for assessing fracture risk are not validated in young women under 40 years and 33 years, respectively. Therefore, individualised risk factors for osteoporosis, such as low BMI and early menopause or FHA, must be considered to guide further investigation and management in accordance with local guidance and protocols. It is reasonable to postulate that several more women in this audit may have undiagnosed osteopenia and/or osteoporosis if screened appropriately.

Conclusion

Whilst we must acknowledge the small numbers in this audit, there is a clear take-home message that bone health is not seriously considered in women with eating disorders. The long-term consequences of this are significant given that women with FHA are at risk of bone loss and/or an inability to reach peak bone mass, which in turn, predisposes them to osteoporosis. A diagnosis associated with a reduced relative survival.²⁷ There is evidence that HRT is superior to the COCP in increasing/maintaining BMD in women with FHA secondary to eating disorders. However, the strength of this evidence is insufficient to clearly guide clinical practice, which is why different advice is given. There appears to be no ‘gold-standard’ for estrogen and progesterone replacement and the duration of therapy is equally ambiguous. In those women with a low BMI, weight gain is arguably the most effective intervention to attenuate or reverse bone loss. However, some studies indicate that both weight restoration and resumption of menstruation are required for BMD to increase.⁶ The HPA axis may remain perturbed even after weight restoration. Therefore, menstrual irregularities may continue in sufferers of eating disorders despite restoration of a ‘normal’ body mass index, highlighting the need for continued therapeutic options. Given the difficulties of treating eating disorders, the judicious prescription of estrogen replacement is a reasonable component of holistic management if the chance of recovery appears remote. Due to a paucity of research, recommendations for estrogen replacement in this clinical context must be based on theoretical knowledge and extrapolated from studies available on HRT in early menopause and POI. With the limited evidence that is available, HRT would appear the superior option. The POISE trial may be a landmark trial for advancing clinical practice in this area.

Recommendations

1. A thorough risk assessment for fragility fracture in all women diagnosed with an eating disorder

2. All patients with FHA should be offered a DEXA scan, and all eumenorrhoeic patients with an eating disorder should be considered for a DEXA scan

3. HRT should be considered a first-line treatment for women with FHA, taking into account individual preferences, desire for contraception, benefits and risks

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Guarantor

H.D.

ORCID iD

Hannah Olivia Davies

References

Roberts

Farahani

Webber

, et al. Current understanding of hypothalamic amenorrhoea. Ther Adv Endocrinol Metab 2020; 11: 204201882094585.

Gordon

Ackerman

Berga

, et al. Functional hypothalamic amenorrhea: an endocrine society clinical practice guideline. J Clin Endocrinol Metabol 2017; 102(5): 1413–1439.

Schneider

. Energy balance and reproduction. Physiol Behav 2004; 81(2): 289–317.

Cartwright

Robinson

Seed

, et al. Hormone replacement therapy versus the combined oral contraceptive pill in premature ovarian failure: a randomized controlled trial of the effects on bone mineral density. J Clin Endocrinol Metab 2016; 101(9): 3497–3505.

Crofton

Evans

Bath

, et al. Physiological versus standard sex steroid replacement in young women with premature ovarian failure: effects on bone mass acquisition and turnover. Clin Endocrinol 2010; 73(6): 707–714.

Steinman

Shibli-Rahhal

. Anorexia nervosa and osteoporosis: pathophysiology and treatment. Journal of Bone Metabolism 2019; 26(3): 133–143.

Prior

. Progesterone for the prevention and treatment of osteoporosis in women. Climacteric 2018; 21(4): 366–374.

Bolland

Grey

Gamble

, et al. Effect of osteoporosis treatment on mortality: a meta-analysis. J Clin Endocrinol Metab 2010; 95(3): 1174–1181.

Armston

Wood

. Hormone replacement therapy (oestradiol-only preparations): can the laboratory recommend a concentration of plasma oestradiol to protect against osteoporosis? Ann Clin Biochem 2002; 39: 184–193.

10.

European Society of Human Reproduction and Embryology . Management of women with premature ovarian insufficiency. In: Guideline of the European society of human reproduction and Embryology [internet]. Brussels, Belgium: European Society of Human Reproduction and Embryology. POI Guideline Development Group, 2015.

11.

National Institute for Health and Care Excellence . Secondary amenorrhoea. London, UK: National Institute for Health and Care Excellence, https://cks.nice.org.uk/topics/amenorrhoea/management/secondary-amenorrhoea/ (2020).

12.

Stevenson

. Prevention and treatment of osteoporosis in women. Post Reprod Health 2023; 29(1): 11–14.

13.

Al-Azzawi

Barlow

Hillard

, et al. Prevention and treatment of osteoporosis in women. Menopause Int 2007; 13: 178–181.

14.

Behary

Comninos

. Bone perspectives in functional hypothalamic amenorrhoea: an update and future avenues. Front Endocrinol 2022; 13: 923791.

15.

DeSapri

Brook

. To scan or not to scan? DXA in postmenopausal women. Cleve Clin J Med 2020; 87(4): 205–210.

16.

Gao

Bagheri

Furuya-Kanamori

. Has the COVID-19 pandemic lockdown worsened eating disorders symptoms among patients with eating disorders? A systematic review. Z Gesundh Wiss 2022; 30(11): 2743–2752.

17.

. Primary osteoporosis in postmenopausal women. Chronic Dis Transl Med 2015; 1(1): 9–13.

18.

Khosla

Oursler

Monroe

. Estrogen and the skeleton. Trends Endocrinol Metabol 2012; 23(11): 576–581.

19.

Davies

Daniels

Upton

, et al. Premature ovarian insufficiency study of effectiveness of hormonal therapy (POISE). In: Hormone therapy for premature ovarian insufficiency: randomised trial and long-term evaluation. London, UK: NIHR, https://poise.ac.uk/documents/poise-protocol-final-2.0-28-jan-2022.pdf (2022).

20.

Sullivan

Sarrel

Nelson

. Hormone replacement therapy in young women with primary ovarian insufficiency and early menopause. Fertil Steril 2016; 106(7): 1588–1599.

21.

Trémollieres

. Contraception orale estro-progestative: quelle différence entre éthinylestradiol et estradiol? [Oral combined contraception: is there any difference between ethinyl-estradiol and estradiol?]. Gynecol Obstet Fertil 2012; 40(2): 109–115.

22.

Seifert-Klauss

Prior

. Progesterone and bone: actions promoting bone health in women. J Osteoporos 2010; 2010: 1–18.

23.

Zhao

J-G

Zeng

X-T

Wang

, et al. Association between calcium or vitamin D supplementation and fracture incidence in community-dwelling older adults: a systematic review and meta-analysis. JAMA 2017; 318: 2466–2482.

24.

Bluming

Tavris

. Estrogen matters. London, UK: Piatkus, 2018.

25.

Hamoda

Panay

Pedder

, et al. The British menopause society & women’s health concern 2020 recommendations on hormone replacement therapy in menopausal women. Post Reprod Health 2020; 26(4): 1–20.

26.

Canonico

Fournier

Carcaillon

, et al. Postmenopausal hormone therapy and risk of idiopathic venous thromboembolism. Results from the E3N cohort study. Arterioscler Thromb Vasc Biol 2010; 30(2): 340–345.

27.

Gregson

Armstrong

Bowden

, et al. UK clinical guideline for the prevention and treatment of osteoporosis. Arch Osteoporosis 2022; 17: 58.