Sexual functioning in women with functional hypothalamic amenorrhea: exploring the relevance of an underlying polycystic ovary syndrome (PCOS)-phenotype

A bidirectional relationship exists between stress response and reproductive function, ultimately affecting also sexual behavior in both sexes [1]. Functional hypothalamic amenorrhea (FHA) is the term used to describe the lack of menstruation not related to organic causes but resulting from various stressors. It is typically associated with three main causes: weight loss, excessive physical exercise, and psychological stress, which can occur alone or, more often, in combination [2]. FHA is responsible for 20–35% of secondary cases of amenorrhea, and about 3% of women with primary amenorrhea [3].

The hallmark of FHA diagnosis is hypogonadotropic hypogonadism related to the suppression of the pulsatile release of hypothalamic gonadotropin-releasing hormone (GnRH) which in turn impairs the pulsatile release of luteinizing hormone (LH) with reduced secretion of 17ß-estradiol (E₂) by the ovary [4]. Numerous hormonal and non-hormonal factors influence the pulsatile secretion of GnRH and come into play in the pathophysiology of FHA, which is rather complex and still not completely understood [2, 4]. Chronic stress is associated with a constellation of neuroendocrine adaptive responses [4, 5] which may influence sexual behavior throughout several common pathways linking hormonal milieu, emotional attitudes and cognitive function [6‐8]. Stressful conditions activate the hypothalamic–pituitary–adrenal axis with an increase in the secretion of corticotropin-releasing hormone (CRH), adrenocorticotropin (ACTH), cortisol, and endogenous opioids. Endogenous opioids play a significant role in the inhibition of the pulsatile release of both GnRH and LH driven by stress [5]. Numerous other hormones and neurotransmitters, including prolactin, dopamine, neuropeptide Y (NPY), GABA, and serotonin negatively affect GnRH release [5]. Reduction of leptin, a 16 kDa adipokine produced by the adipose tissue, also contributes to the suppression of pulsatile GnRH release in patients with FHA [9]. However, GnRH neurons do not express leptin receptors and therefore other peptides, including kisspeptin, which is the most potent stimulator of GnRH secretion, mediate leptin regulation of hypothalamic GnRH release [10]. Kisspeptin acts directly on hypothalamic GnRH neurons via the kisspeptin receptor and causes GnRH release into the hypophyseal-portal blood circulation. Kisspeptin neurons co-express neurokinin B and dynorphin (KNDy neurons), which act autosynaptically to synchronize the pulsatile secretion of kisspeptin with excitatory (neurokinin B) and inhibitory signals (dynorphin A). Kisspeptin plays a main role in the regulation of reproduction but also of sexual function and other aspects of reproductive behavior from rodents to humans [11]. Hypoestrogenism, the final hormonal consequence of the impairment in GnRH and gonadotropin pulsatile release, may inflect central and peripheral components of sexual function [12]. Hypoandrogenism that becomes evident in severe forms of hypothalamic amenorrhea (HA) [13] may also play a role. Dundon and colleagues [14] reported that women with FHA had more sexual problems than healthy controls. Moreover, psychologic symptoms that contributed to the onset of FHA [7] partially mediated the relationship between FHA and sexual functioning [14].

FHA may coexist with polycystic ovary syndrome (PCOS) [2], another very common cause of hormonal infertility in women of reproductive age, affecting up to 15% worldwide, depending on the diagnostic criteria [15]. Wang and Lobo [16] explored the spectrum of FHA and polycystic ovarian morphology (PCOM) and identified women with FHA and a PCOS-phenotype that may have inherently hyperandrogenic ovaries quiescent because of low gonadotropin secretion resulting from hypothalamic functional suppression. Recent considerations suggest a better distinction between FHA and PCOS is relevant to improve treatment strategies for anovulatory infertility [17]. Indeed, a correct assessment has an important significance both in terms of disease severity and individualized care of FHA based on the hormonal milieu and increased follicle number per ovary [18] without the typical peripheral follicular distribution, the so-called multicystic ovarian morphology (MCOM) [19]. In spite of a real consensus on the ultrasound criteria to define women with FHA carrying PCOS-like ovaries [18, 20], it seems likely they display a peculiar hormonal profile with an elevation of some androgen levels [21‐23]. These hormonal features may affect individual well-being, including mood and sexual functioning and define different phenotypes of women with FHA.

That being so, the present retrospective study aimed to investigate sexual function and distress in women with FHA as compared to women with FHA + PCOS-phenotype. Psychometric variables were also measured to look over emotional milieu of these two groups of patients. In addition, we explored the relationship between sexual function and distress with the hormonal profile of FHA women.

This is a retrospective study conducted on women referred to the Unit of Gynecological Endocrinology at IRCCS San Matteo Foundation of the University of Pavia, in northwest Italy, for diagnosis and treatment of menstrual dysfunction. The local ethics committee granted approval for the analysis of medical records of those patients who have signed an informed consent.

The following inclusion criteria were established: age ≥ 16 years (at least 2 years since menarche), body mass index (BMI) ≥ 16 kg/m², the disappearance of menses for at least 3 months before clinical evaluation, low or normal LH levels, 17ß-estradiol (E₂) levels ≤ 100 pg/ml, and one or more recent predisposing factors (weight loss, excessive physical activity, stressful situations).

The following exclusion criteria were used: pregnancy and lactation, any hormonal therapy (including combined hormonal contraceptives), history of major diseases including endocrine and psychiatric disorders, psychoactive medications, tobacco or illicit drug use, alcohol dependence, overt eating and sleeping disorders.

To serve the scope of investigating sexual functioning, we further excluded those women who did not report a sexual debut and declared not being in a stable sexual relationship in the preceding three months. Finally, we included 79 women with FHA in the present analysis.

According to the Endocrine Society Clinical Practice guideline for the diagnosis and treatment of FHA [2], we reconsidered the records of our study sample to identify a potential PCOS-phenotype, based on menstrual history, clinical signs, and two-dimensional transvaginal ultrasound evaluation. Those FHA women with MCOM, as previously described (six or more small follicles in one plane distributed within normal stroma) [19], were classified as FHA + PCOS-phenotype (n = 43), whereas the remaining women as typical FHA (n = 36). Endometrial thickness was lower than 4 mm in every woman.

Information about age, age at menarche, education and body mass index (kg/m²) were included in the present analysis. Bodyweight (kg) and height (m) were measured in a fasting state, without shoes and without wearing heavy clothes, by using the same calibrated scale and stadiometer. We considered available hormones routinely measured in women with FHA at the first consultation. Assays validated as standard practice in our University Research Hospital were the following: follicle-stimulating hormone (FSH) [follicular phase range 2.8–11.3 IU/L], luteinizing hormone (LH) [follicular phase range 1.1–11.6 IU/L], 17ß-estradiol (E₂) [follicular phase range 15–160 pg/ml], prolactin [normal range 1.9–25 ng/ml], thyroid-stimulating hormone (TSH) [normal range 0.4–4.0 mIU/L), free triiodothyronine (FT3) [normal range 1.8–4.2 pg/ml], free thyroxine (FT4) [normal range 8.0–19.0 pg/ml], androstenedione [normal range 0.46–3.39 ng/ml], total testosterone (TT) [normal range: 0.15–0.80 ng/ml], dehydroepiandrosterone sulfate (DHEAS) [normal range 0.35–4.3 mcg/ml], cortisol [normal range AM 4.3–22.4 mcg/dl], and insulin [normal range 3–25 mIU/L]. FSH, LH, E2, prolactin, androstenedione, Total T, DHEAS, cortisol were measured by chemiluminescence (SIEMENS Immulite 2000, Germany). The minimal detectable limit for each hormone was as follows: 0.1 IU/L (FSH and LH), 15 pg/ml (E2), 0.5 ng/ml (prolactin), 0.3 ng/ml (androstenedione), 0.15 ng/ml (TT), 0.03 mcg/ml (DHEAS), and 0.2 mcg/dl (cortisol). The inter-assay coefficients of variation were 4.4% (FSH), 6.1% (LH), 7.3% (E2), 5.8% (prolactin), 6.2% (androstenedione), 6.6% (total T), 5.9% (DHEAS), and 6.1% (cortisol). The intra-assay coefficients of variation were 3.8% (FSH), 4.9% (LH), 6.1% (E2), 5.1% (prolactin), 5.5% (androstenedione), 5.5% (total T), 5.1% (DHEAS), and 5.2% (cortisol). Insulin was measured by the ADVIA XPT (Siemens Healthcare; detection limit 0.5/L, inter-assay coefficients of variation 5.9%, intra-assay coefficients of variation 4.6%). TSH (third generation), FT4 and FT3 assays were performed on a fully automated ADVIA Centaur analyzer (Siemens Healthcare Diagnostics) with a sensitivity of 0.02 mIU/L (TSH) and 1.1 pg/ml (FT3, FT4), an inter-assay coefficients of variation of 4.9% (TSH), 4.2% (FT3) and 4.4% (FT4) and an intra-assay coefficients of variation of 3.9% (TSH), 3.6% (FT3) and 3.8% (FT4).

Information deriving from validated psychometric questionnaires to evaluate sexual functioning, as well as other typical intra-personal characteristics of FHA (body image, bulimic risk, mood, and personality), were available and listed as follows:

The present study demonstrated that women with FHA have a very high rate of sexual symptoms as part of their menstrual dysfunction, especially those with typical FHA as compared to FHA-PCOS-phenotype. Sexual distress was present to a lower extent in both groups, without any significant difference, in line with similar levels of sexual satisfaction. Interestingly, a potential role of sex hormones, mainly LH-driven androstenedione, in influencing some components of women’s sexual response (desire, arousal, lubrication, and orgasm) emerged, with high sexual pain scores in both FHA groups, regardless of circulating E₂ or androgen levels.

Typical FHA and FHA + PCOS phenotypes were superimposable from a psychosocial standpoint evaluated by body attitude, state and trait anxiety, depression, bulimic risk, and perfectionism. Our psychometric data were consistent with the neuropsychological profile of FHA adolescent [32] and young adult [7] women reported in previous studies and confirmed that typical FHA and FHA + PCOS-phenotype represent a continuum on a spectrum of menstrual dysfunction. That being so, the peculiar psychoneuroendocrine environment of women with FHA [33] is likely to play a fundamental role in sexual functioning in both groups. However, our attempt to compare typical FHA with FHA + PCOS-phenotype outlined a greater impairment in sexual function of those women with more profound hypogonadism. Whether abnormalities of sex hormones may significantly contribute to psychosexual well-being of women with menstrual dysfunction awaits further evidence in prospective studies with sensitive assays. In a cross-sectional study, circulating androgens (total and free testosterone, androstenedione, and DHEAS), measured by mass spectrometry, correlated to the FSFI sexual desire domain score, especially in a subgroup of women aged 25–44 years [34]. Given the multidimensionality of sexual functioning, controversies in linking levels of androgens to specific domains of the sexual response still exist in many gynecological conditions, including PCOS [35]. A recent meta-analysis including 34 studies and 3268 women with a mean age of 36.5 years showed that total testosterone was associated with better sexual function [SMD = 0.55 (0.28;0.82), p < 0.0001] [36]. FHA is among the potential causes of low testosterone due to anovulation, possibly leading to HSDD [37]. Moreover, both hypoestrogenism [38] and hypoandrogenism [39] may contribute to the clinical manifestation of sexual symptoms associated with genital involution and poor responsiveness. However, two subsequent consensus publications of the International Society of Sexual Medicine (ISSM) recognized the paucity of research on the potential impact of FHA and its treatment on sexual functioning [40, 41]. An early study [42] explored the possibility that impaired sexual function may result from reduced levels of testosterone in subjects with secondary amenorrhea of hypothalamic origin diagnosed according to the presence of risk factors (i.e., weight loss before the onset of amenorrhea, low body weight, strenuous exercise, and vegetarianism). Eight women with FHA associated with these particular life-styles demonstrated impaired sexual function and significantly lower circulating testosterone levels as compared with 8 normally menstruating women [42]. Interestingly, women with FHA were asked to produce erotic fantasies demonstrating a reduced capacity for sexual fantasizing, less subjective sexual excitement, and less vaginal vasocongestion as compared to healthy women. Treatment with testosterone undecanoate (40 mg) increased vaginal vasocongestion in women with FHA without affecting the subjective sexual experience in response to the exposure to an erotic movie [42].

Sex hormone-binding globulin (SHBG) has been previously reported as a promising parameter to distinguish between FHA and PCOS [18]. Indeed, SHBG levels were lower in PCOS patients, regardless of BMI [18]. Moreover, SHBG has been related to sexual function because oligomenorrheic and/or hirsute women with values lower than 33.4 nmol/l displayed adverse metabolic profiles along with higher clitoral pulsatility index (PI), an index of vascular resistance [43]. These results agreed with recent literature reporting a higher clitoral PI in patients with metabolic impairment and sexual dysfunction [44‐46]. Results from a substudy of the Grollo-Ruzzene cross-sectional study, which recruited 761 premenopausal women from eastern states in Australia, showed a negative association between sexual desire and SHBG, in line with other studies conducted on premenopausal women [47]. On the other hand, a recent cross-sectional study showed a positive correlation between SHBG and total FSFI score (r = 0.39; p = 0.02) in 36 postmenopausal women (age range 45–65 years) [48]. Of note, SHBG was positively correlated with three specific FSFI domains [orgasm (r = 0.39; p = 0.01), satisfaction (r = 0.33; p = 0.04), and pain (r = 0.44; p < 0.01)]. Collectively, these findings suggest the role of SHBG in modulating female sexual function merits further investigation. Unfortunately, SHBG data were not available in our present study. Among factors that can influence SHBG levels, thyroid hormones played a significant role. In our sample, levels of FreeT3 and FreeT4 were statistically significantly lower in women with typical FHA compared to women with FHA + PCOS phenotype. It is a matter of fact that both hypothyroidism and hyperthyroidism are associated with changes in concentrations of SHBG, testosterone, and E₂. However, the direct relationship between thyroid function and sexual function is still not clear. Indeed, a recent study reported that, although genetically predicted thyroid function was associated with sex hormones, it was not associated with ovulatory function in women (duration of menstrual cycle, age at menarche and menopause and reproductive lifespan) and erectile dysfunction in men [49].

In our study sample, less than half of women with FHA was clinically distressed by sexual symptoms and only a few of them reported low sexual desire associated with sexual distress (HSDD). These data supported previous evidence of similar satisfaction with a sexual relationship with respect to healthy women [14]. Sexual symptoms were confirmed to co-occur in FHA women more vulnerable to mood disorders [14] as it was found in general clinical practice [50]. Even in a nonclinical sample of young women, depression-specific anhedonia and sexual desire were closely related [51]. Moreover, in the same study physiological hyperarousal linked to anxiety gave rise to sexual arousal difficulties and vaginal pain [51]. In women with FHA, the common link might be the increased activity of hypothalamus–pituitary–adrenal axis (HPA) associated with the impairment of some central pathways (opioids and serotonin) [52], which play a fundamental role in the neuroendocrine balance driving sexual functioning [6] and also contribute to the onset of menstrual dysfunction [4]. On the other hand, other common pathways connecting the areas of emotion dysregulation/impulsivity and perfectionism/over-control both with eating disorder-specific psychopathology and with different types of sexual impairment [53] might explain the high rate of sexual symptoms in women with FHA, including sexual pain, reported in the present study.

The major limitations of our study are the retrospective design and the lack of sensitive assays to measure androgens. On the other hand, the present study has the strength that it was conducted by using validated instruments encompassing several psychosexual dimensions in a clinically and hormonally well-characterized sample of women with FHA. Being aware of the paucity of data investigating sexual functioning in amenorrheic women, we aimed to present real-life data to guide the daily practice of every clinician dealing with anovulatory menstrual disorders. Whether the experience of infertility associated with the absence of menstrual periodicity contributes to the manifestation of sexual symptoms is presently unknown and it awaits further investigation.

Overall, our findings underscored the need to include sexual counselling in the diagnostic evaluation for FHA. In addition, appropriate management should provide not only cognitive-behavioral therapy [54] to overcome risk factors, such as mood disorders, dysfunctional attitudes, eating disorders, body image issues, excessive exercise, and other stressors, but also sexual symptoms. Tailored hormonal strategies should be investigated also in the context of sexual care.

In conclusion, FHA is a suitable paradigm to investigate the complexity of sexual functioning from both a neuroendocrine and a psychosocial standpoint. Along with many other stressors, we cannot exclude that a high rate of sexual symptoms may contribute to the maintenance of such menstrual dysfunction. Moreover, the recognition of an underlying PCOS-phenotype can be helpful to investigate discrete components of sexual functioning possibly linked to different profiles of hypogonadism. Prospective studies are warranted to determine the relevance of our retrospective data in the clinical management of women with FHA.