Interpretation
Our data supports the strength of known FHA risk factors including, high-level cardiovascular exercise, psychological stress19 and weight loss preceding amenorrhoea.20-22 Fat loss and dieting decrease circulating leptin; hypoleptinemia reduces stimulation of hypothalamic neuronal networks, thus suppressing GnRH secretion.
23,24 Despite increased exercise states, however, 21/50 and 8/50 of women with FHA who a received DEXA scan, were osteopenic and osteoporotic, respectively. We observed lower rates of reduced BMD, than those in the literature; with reported rates of osteopenia and osteoporosis as high as 83% and 33% respectively in women with FHA
25,26;this may have been attributed to the low rates of DEXA scanning in our study group. 50% of women with FHA reported a positive ED history, which was comparable to previously reported rates of 55%. 27
A large retrospective study investigating the potential for misdiagnosis of FHA found that 86% of lean women with FHA would have met the Rotterdam criteria for PCOS diagnosis, had FHA not been detected.
17 Although we have highlighted the pertinent clinical features of FHA, it is important to recognise that the underlying aetiology of FHA often involves a varying degree of interacting factors and there remains a lack of consensus on what constitutes excessive exercise, significant nutrition restriction or psychological stress. This is deemed problematic in the absence of reliable investigative or biochemical diagnostic tests and the growing concerns around misdiagnosing women with FHA with PCOS.
An important finding of our study was the significantly lower BMI in women with FHA compared to those with PCOS, this is supported by previous reports.28 Whilst BMI was found to best distinguish between FHA from PCOS extremely effectively, the large range reported emphasises the importance of not ruling out PCOS or FHA based on BMI alone. The heterogenous and often insidious presentation of FHA calls for a comprehensive clinical approach, which includes combining clinical parameters with the patient’s history (occupational, exercise and nutrition) to first suspect and second, diagnose FHA. A recent study by Rodino et al demonstrated 65% of fertility specialists did not screen for eating disorders during preconception assessments, despite 83.7% of fertility specialists in their study agreeing it was important to do so, this was perhaps explained by the fact that only 37.5% felt confident in their ability to recognise symptoms of an eating disorder.29 This is of particular significance given that 76.4% to 100% of patients do not disclose their eating disorder.30,31In their retrospective study, Bradbury et al argue that lack of a detailed diet and exercise history in patients with a low BMI may result in them being misdiagnosed with PCOS.
17
Similarly, to our biochemical findings, one study comparing women with FHA with PCOS, found lower baseline LH and estradiol in women with FHA, with no difference in FSH levels.32 Low estradiol levels are typical of FHA, secondary to suppression or insufficient stimulation of the HPO axis, in which no anatomic or organic pathology can be identified. 33 Elevated LH and mild FSH suppression characterises PCOS 10 whereas, whilst both gonadotropins are reduced in FHA, LH is suppressed more significantly.34,35 This likely explains why estradiol and LH were highly effective at distinguishing FHA from PCOS and why FSH was a poor distinguisher. The highly significant difference in LH between FHA and PCOS combined with the similar FSH levels meant that the LH:FSH ratio was also good at distinguishing the two conditions. Knowledge of the patient’s baseline LH and FSH levels is important in the context of diagnosis but also when considering fertility treatment options. For example, the use of clomiphene citrate is unlikely to be recommended, owing to the underlying hypoestrogenic state in FHA patients.36 Nonetheless the use of exogenous gonadotrophin therapy will likely require both LH and FSH stimulation37; this would require patients to be educated on the risks of ovarian hyperstimulation syndrome (OHSS), especially in the presence of low BMI which increases sensitivity to gonadotrophin stimulation.38