Functional hypothalamic amenorrhoea: a diagnosis of exclusion

Functional hypothalamic amenorrhoea is common but often misdiagnosed, risking inappropriate management and compromised patient education and counselling

Secondary amenorrhoea before the age of 45 years, not due to pregnancy or lactation, lasting over 3 months in a previously eumenorrhoeic woman, or 6 months if menses have been irregular,1 is a cause for concern. Prolonged oestrogen deficiency due to amenorrhoea in reproductive aged women is associated with bone loss, increasing fracture susceptibility,2 dyslipidaemia3 and endothelial dysfunction, potentially increasing cardiovascular disease risk.4,5 Additionally, there is the psychological impact of anxiety about infertility and, conversely, possible unexpected ovulation and undesired pregnancy.6 Women with prolonged amenorrhoea experience more sexual dysfunction, depression and anxiety compared with women with eumenorrhoea.7 In this article, we summarise the available literature on the diagnosis and management of functional hypothalamic amenorrhoea (FHA). We searched PubMed, Google Scholar and the Cochrane Library for publications pertaining to the diagnosis and management of FHA from inception to June 2021. A hypothetical case study (Box 1) illustrates how the available literature can guide clinical practice.

The prevalence of secondary amenorrhoea remains uncertain. Frequent causes include FHA, polycystic ovary syndrome (PCOS), hyperprolactinaemia, premature ovarian insufficiency and early menopause (< 45 years of age). Less common aetiologies to consider and exclude include congenital adrenal hyperplasia, while rare causes include hypothalamic hypogonadism and androgen secreting tumours. Consequently, secondary amenorrhoea requires a thorough evaluation to identify the cause, as indicated in Box 2 and Box 3. Most importantly, a careful history that covers psychosocial and medical components, as described in Box 1, usually provides the key to the diagnosis.

FHA is chronic anovulation with no underlying anatomical or pathological cause, and is therefore a diagnosis of exclusion.2 PCOS, which affects an estimated 7–12% of premenopausal women, requires at least two of oligo/amenorrhoea, hyperandrogenism and sonographic evidence of polycystic ovaries for a diagnosis.8 Premature ovarian insufficiency is complete cessation of ovarian function before the age of 40 years,9 and has a reported prevalence of 4%.10 Premature ovarian insufficiency can be considered after at least 4 months of amenorrhoea with elevated follicle‐stimulating hormone (FSH) levels > 25 mIU/mL on two occasions at least 4 weeks apart.11 Although not required to make the diagnosis, low oestradiol levels characterise this condition. Hyperprolactinaemia, which may present as amenorrhoea with or without galactorrhoea, accounts for about 14% of secondary amenorrhoea in women aged 21–30 years.12 Importantly, hyperprolactinaemia may be secondary to a number of medications, notably antipsychotics and antidepressants,13 as well as clinical and subclinical hypothyroidism.14

FHA and PCOS are the two most common causes of hormonal subfertility and they can be difficult to differentiate when the classical clinical features of these two distinct conditions are absent. Such difficulty is encountered in lean women without pronounced hyperandrogenism, disturbed eating behaviour, weight loss or substantial exercise.15 While oestradiol levels are mostly low in FHA and in the normal range in PCOS, this is not always the case. Moreover, women with FHA often have multifollicular ovaries.16,17 Therefore, neither a high ovarian follicle count on ultrasound nor a high blood anti‐Müllerian hormone concentration can reliably differentiate PCOS from FHA.15,16 However, low sex hormone‐binding globulin is pathognomonic for PCOS,18 with a normal to high concentration indicating PCOS to be an unlikely diagnosis.18,19 Taken together, a pelvic ultrasound will not aid the diagnosis for normal weight or lean women with secondary amenorrhoea, who do not have biochemical or clinical hyperandrogenism. A progestogen challenge test can be used to induce a withdrawal bleed as evidence of oestrogen exposure, as occurs in women with PCOS, and no uterine outflow obstruction. As progestogens are often poorly tolerated and a bleed would be unexpected in women with chronically low oestradiol levels, this would not be diagnostically informative in our described patient (Box 1). The key differential will be establishing a psychosocial profile consistent with FHA.15

Hypothalamic amenorrhoea occurs as a result of disordered frequency and amplitude of hypothalamic gonadotrophin‐releasing hormone (GnRH) secretion, leading to loss of cyclical production of the gonadotrophins, FSH and luteinising hormone, and subsequent anovulation.20 For most women, however, serum FSH insufficiency is subtle, with levels remaining in the low normal range.17 Studies in humans have revealed that GnRH release is centrally controlled by a complex interplay between the neurotransmitters kisspeptin, neurokinin B and dynorphin released by KNDy neurons in the infundibular nucleus, and kisspeptin from the preoptic area, as shown in Box 4.21 Kisspeptin, which stimulates GnRH neurons in the hypothalamus, is inhibited by dynorphin and cortisol but stimulated by neurokinin B.22 Perturbations in this system can occur in response to physiological and/or psychological stress. Hence, mildly elevated 24‐hour mean cortisol levels are seen in women with FHA, which reflects a physiological response to stress.22 While we tend to characterise women with FHA as undertaking intense exercise, having a caloric deficit or marked weight loss, not all women with FHA will have an overt eating disorder or substantial recent weight loss.22 The aetiology is frequently subtle. Consequently, women with FHA can rarely identify a solitary stressor, as FHA often results from a synergistic combination of psychogenic and metabolic stressors, which can be seemingly minor.22 Thus, the hormonal changes that result in FHA can be precipitated by minimal stressors, such that FHA may develop without a clear history of exercise excess, disordered eating or significant weight loss, and may go undetected in weight stable, non‐exercising women.

The treatment of FHA involves explaining to the affected woman the role of psychological and metabolic stressors in hypothalamic–pituitary–adrenal axis disruption, and working with them to identify relevant modifiable behaviours.23 It is vital to reassure women with FHA that the condition is reversible and that fertility will be restored with resumption of menstrual cycles, or easily triggered with available interventions.2 Addressing FHA mostly requires a multidisciplinary team approach that may include expert medical, dietetic and psychological care. Psychological stressors should be sought and managed in all instances, with management of disordered eating and exercise excess where relevant.2 All women should receive a nutritional assessment, as nutritional deficiency may occur as a result of dietary preferences, such as vegetarianism or veganism (iron and zinc insufficiency), or dairy exclusion (calcium insufficiency). Cognitive behaviour therapy has been shown to lead to recovery of ovarian function, through addressing problematic behaviours and facilitating development of healthy coping mechanisms.24 Trauma therapy and other psychological support approaches should be offered where indicated.

Although FHA may be transient, it has been demonstrated to have lasting impacts, including low bone mineral density25 and an increased association with angiographic coronary artery disease.26 Bone mineral density and vitamin D levels should be measured in all women who have had 6 months of amenorrhoea, or less if nutritional depletion is likely to have preceded the amenorrhoea. As bone loss may commence after 6–12 months of amenorrhoea, hormone therapy in the form of oestrogen replacement with cyclic or continuous progestogen therapy should be commenced, to prevent bone loss if menstruation has not resumed within 6 months of appropriate intervention.

The US Endocrine Society guideline recommends against the use of a combined oral contraceptive pill to provide oestrogen replacement because, as intended, this inhibits ovulation and therefore masks recovery.2 Furthermore, oral oestrogen, whether as ethinyloestradiol or oestradiol, has not been found to improve bone mineral density in women with FHA.27 Conversely, transdermal oestradiol (100 µg 17‐β oestradiol) with cyclic progesterone has been shown to improve bone mineral density in adolescents with anorexia nervosa28 and in athletes aged 14–25 years with oligoamenorrhoea.29 Women with FHA often have disturbed cortisol and thyroid hormone production, which also affects their bone health.23,30 It is not known whether oral oestrogen further interferes with cortisol or thyroid hormone homeostasis through its effects on hepatic production of cortisol and thyroid‐binding globulins, thus impairing the efficacy of oral oestrogen on bone recovery.30 Therefore, specifically in the management of FHA, first line therapy should be with transdermal oestradiol (which does not undergo first pass hepatic metabolism) to optimise bone health,2,27 unless contraception is required, in which case an oestradiol‐containing combined oral contraceptive pill is preferred. A progestogen should be co‐prescribed to protect against endometrial hyperplasia and cancer. While body‐identical progesterone is the most physiological option and is preferred, many women will be limited to a synthetic progestin by cost. Although the goal is to achieve a transdermal oestradiol dose (with a patch or a gel) equivalent to a 100 µg/day oestradiol patch, treatment should be initiated at a lower dose to avoid side effects such as mastalgia and heavy bleeding, which many women with FHA may not tolerate. The US Endocrine Society recommends against the use of bisphosphonates for bone protection in FHA, as no studies have provided evidence for their use in this condition.2 Moreover, bisphosphates are stored in bone for many years and should therefore preferably not be used in women likely to subsequently become pregnant.2 Other antiresorptive agents have not been sufficiently studied in premenopausal women to justify their use.

In summary, FHA, although a common cause of secondary amenorrhoea, is a diagnosis of exclusion. Given the potential adverse physical health outcomes associated with secondary amenorrhoea due to any cause, combined with the psychological consequences of delayed diagnosis as well as misdiagnosis, it is important that women with secondary amenorrhoea are fully assessed when they first present. This should include assessment of bone health and cardiovascular health as part of the initial work‐up and management. A proactive approach to secondary amenorrhoea will enable a diagnosis to be made promptly and accurately, and expedite targeted treatment of the underlying cause.

Box 3 – Approach to the work‐up of secondary amenorrhoea

---

ACTH = adrenocorticotropic hormone; FSH = follicle‐stimulating hormone; FT4 = free thyroxine; hCG = human chorionic gonadotrophin; IGF‐1 = insulin‐like growth factor‐1; MRI = magnetic resonance imaging; PCOS = polycystic ovary syndrome; POI = premature ovarian insufficiency; TSH = thyroid‐stimulating hormone.

Box 4 – Central control of female reproduction

---

Kisspeptin, released from neurons in the preoptic area and infundibular nucleus, triggers gonadotrophin‐releasing hormone (GnRH) release. This is inhibited by dynorphin (DYN) via its receptor, the κ opioid receptor (KOR) and stimulated by neurokinin 3B (NK3B) and neurokinin 1B (NK1B) via receptors in the KNDy neuron complex. The pattern of kisspeptin release translates into the frequency and amplitude of GnRH pulses, which determine follicle‐stimulating hormone (FSH) and luteinising hormone (LH) release from the anterior pituitary gland.