Eating disorders are life-threatening conditions that disproportionately affect females, often during child-bearing years. Although the endocrinological and reproductive sequelae of these conditions often fall within the treatment purview of obstetrician-gynecologists, the assessment of eating pathology is challenging and often not part of routine clinical care. This commentary focuses on one of the common presenting symptoms of eating disorders in women, menstrual dysfunction, and discusses considerations for its clinical management in gynecology. Assessment of menstrual status provides a natural starting point for provider-patient discussion of disordered eating and weight behavior. Routine screening for eating disorders is critical and must be universal given the serious long-term consequences of these disorders.
There is substantial evidence that eating disorders (anorexia nervosa, bulimia nervosa, and binge eating disorder) often go unrecognized by health care providers. Data collected from physicians working in obstetrics and gynecology suggests that although the majority of providers assess body weight and exercise habits, less than half assess a history of eating disorders or for concerns relating to body image.
An assessment of specific eating disorder behaviors such as binging and purging are rare, with less than 10% of those surveyed indicating that such as assessment was part of their routine clinical care. Thus, it is not surprising that physicians in these setting recognize less than 1 in 10 cases of bulimia nervosa or binge eating disorder. Low rates of eating disorder assessment represent a missed opportunity. Early identification and intervention is critical, given the cumulative negative impact of these disorders on mental and physical health. Longer latency to treatment is associated with poor prognosis, including increased mortality.
Although early identification is paramount, several barriers toward recognition exist including lack of provider training, diffusion of responsibility, and patient denial of illness. In a study by Leddy et al surveying Fellows of the American College of Obstetricians and Gynecologists, more than one third of physicians reported that training in diagnostic assessment of eating disorders was nonexistent during their residency; only 4% rated their training as adequate or greater. Moreover, despite the frequent neuroendocrine and reproductive sequelae of these disorders, only half of those surveyed reported that assessment of eating disorders was part of their role as providers.
Barriers on the provider side are compounded by the fact that most individuals with eating disorders do not seek treatment. Eating disorders, particularly anorexia nervosa, are often ego syntonic; denial of illness and ambivalence toward treatment are common. Furthermore, patients may be reluctant to disclose their symptoms because of the shame and stigma attached to these disorders. Increasing evidence also points to poor mental health literacy as well as the social acceptability of fasting, extreme exercise, or other nonpurging weight-control methods as contributing factors. The frequent reticence among those with eating disorders to seek medical or psychiatric care coupled with poor mental health literacy makes improving provider training around eating disorder recognition critical.
This commentary focuses on one of the common presenting symptoms of eating disorders, menstrual dysfunction, as a starting point for provider-patient discussion and highlights the critical importance of obstetrician-gynecologists in providing collaborative care. The impact of eating disorders on pregnancy, obstetric, and perinatal outcomes has been discussed extensively elsewhere ; therefore, the focus here is on introducing one method for eating disorder identification to provide guidelines for routine clinical management of nonpregnant women with eating disorders.
Clinical definitions
Anorexia nervosa is characterized by the restriction of food intake resulting in a significantly low body weight and an intense fear of weight gain (or persistent behavior interfering with weight gain) despite this low weight as well a disturbance in the way in which body shape/weight are experienced. Determination of significantly low body weight is made in the context of age, sex, developmental trajectory, and physical health, though Diagnostic and Statistical Manual of Mental Disorders , fifth edition, provides body mass index guidelines for illness severity: ≥17 kg/m 2 (mild), 16–16.99 kg/m 2 (moderate), 15–15.99 kg/m 2 (severe), <15 kg/m 2 (extreme).
Anorexia nervosa is further subtyped depending on the presence (binge eating/purging type) or absence (restricting type) of regular engagement in binge eating and/or purging behavior. Diagnostic crossover between anorexia nervosa subtypes is common. The Table presents physical symptoms commonly associated with anorexia nervosa.
| Eating disorder | Clinical characteristics | Commonly associated physical symptoms |
|---|---|---|
| Anorexia nervosa a | Restrictive eating May also binge eat and/or purge Underweight/emaciated | Amenorrhea Osteoporosis Bradycardia Hypotension Hypothermia Lanugo hair Dry skin, brittle nails Gastric dilation, constipation Electrolyte abnormalities Edema |
| Bulimia nervosa | Binge eating with purging, fasting or other compensatory behavior Normal weight | Menstrual irregularities PCOS Parotid enlargement Dental erosion Russell’s sign (scarring on back of hand due to vomiting) Hypotension Edema Electrolyte abnormalities |
| Binge eating disorder | Binge eating in the absence of compensatory behavior Typically overweight or obese | Menstrual irregularities PCOS Medical complications associated with obesity (heart disease, diabetes, hypertension, etc) |
a If purging associated physical symptoms for BN also apply.
Bulimia nervosa is characterized by recurrent episodes of binge eating, or uncontrolled eating of an abnormally large amount of food in a discrete time period, that is coupled with inappropriate compensatory behavior (eg, self-induced vomiting, laxatives misuse, diuretics, enemas or other medications, fasting, or excessive exercise) to prevent weight gain. For diagnosis, these behaviors must occur at least once per week on average; however, the frequency increases to multiple times per day in severe cases. Similar to anorexia nervosa, self-evaluation is unduly influenced by body shape and weight. Individuals with bulimia nervosa are typically average or slightly above average weight. Thus, weight is the primary symptom differentiating bulimia nervosa from binge eating/purging-type anorexia nervosa ( Table ).
The most prevalent eating disorder, binge eating disorder, is characterized by binge eating that occurs on average at least once per week in the absence of compensatory behavior. Episodes of binge eating are associated with significant distress and often feelings of disgust, shame, or embarrassment. Binge eating often occurs alone and is marked by eating more rapidly than normal, until uncomfortably full or when not physically hungry. The majority of individuals with binge eating disorder are overweight or obese, and binge eating disorder is common among those seeking weight-loss surgery. Thus, it is important to note that eating disorders occur in individuals of all weight ranges.
Menstrual disturbance: a common presenting symptom
Endocrine and physiological abnormalities including primary and secondary amenorrhea and menstrual dysfunction are commonly associated with eating disorder pathophysiology. Although no longer a requirement for diagnosis, approximately 90% of women with anorexia nervosa report amenorrhea, the majority of which report secondary amenorrhea. Although there is a clear association between reduction of body weight, nutritional status, and amenorrhea, some women with low body weight have normal or infrequent menses.
Amenorrhea also precedes weight loss in approximately 20% of women with anorexia nervosa. Importantly, nutritional status, not psychiatric status, seems to be the determining factor that differentiates between women with and without amenorrhea. Studies report few meaningful differences in the course of illness, psychiatric comorbidity, eating disorder-related cognitions, or treatment outcome between menstruating and nonmenstruating patients. Thus, the absence of a regular menstrual cycle can be thought of as one marker of nutritional insufficiency but not pathognomonic for the disease.
Amenorrhea associated with anorexia nervosa is secondary to hypothalamic dysfunction, characterized by the suppression of gonadotrophin-releasing hormone pulsatility and subsequent estrogen deficiency. This malnutrition-induced functional hypothalamic amenorrhea is associated with a host of other physiological changes reviewed extensively elsewhere. These include a sick euthyroid pattern seen in chronic starvation (low endogenous levels of triiodothyronine, whereas reverse triiodothyronine concentrations and thyroxine to triiodothyronine ratio values are elevated), overactivity of the hypothalamic–pituitary-adrenal axis (increased secretion of corticotrophin-releasing hormone, hypercortisolemia), and abnormalities in appetite-regulating hormones (hypoleptinemia, elevated peptide YY).
Hypoleptinemia signaling a state of energy deficiency may mediate the neuroendocrine changes seen in hypothalamic amenorrhea. Eumenorrheic women with anorexia nervosa have higher serum leptin levels and more body fat compared with ammenorheic women of similar weight, suggesting that fat mass may play a role in persevering menstrual function. Although leptin administration has been shown to normalize gonadotrophin-releasing hormone pulsatility and induce menstruation in normal-weight women, there is no established threshold by which leptin levels can predict return to or cessation of menstruation. The considerable overlap in serum leptin levels between menstruating and nonmenstruating women with anorexia nervosa makes this unlikely and underscores the notion that leptin is likely one of many factors implicated in the pathophysiology of hypothalamic amenorrhea in women with anorexia nervosa.
Although most frequently recognized in anorexia nervosa, menstrual dysfunction is a frequent presenting symptom for women across the eating disorder diagnostic spectrum including those with bulimia nervosa, binge eating disorder, and other specified feeding or eating disorders. In a study of 1705 women with eating disorders, approximately 40% of women with bulimia nervosa presented with amenorrhea and 19% reported oligomenorrhea.
Menstrual irregularities were also common (49%) in those with other specified presentations. Specific eating disorder symptoms show a distinct relationship to patterns of menstrual dysfunction. Although severe caloric restriction, excessive exercise, and low lifetime body mass index are commonly associated with the cessation of menses, binge eating and/or purging are associated with irregular or normal menstruation. Binge eating may provide sufficient, albeit irregular, caloric absorption to preserve the menstrual cycle to some extent. Conversely, a high frequency of vomiting may interfere with the menstrual cycle by increasing dopaminergic and opioid activity. Moreover, there is a noteworthy association between bulimia nervosa, binge eating disorder, and polycystic ovary syndrome, which itself is marked by menstrual irregularities and disruption.
Large fluctuations and dietary intake brought on by binging, restriction, and/or purging may facilitate the expression of morphologically polycystic ovaries, potentially mediated by abnormalities in insulin secretion. Ultrasonography has revealed abnormal ovarian morphology in bulimia nervosa, with greater than 75% of patients having polycystic ovaries. In addition, 1 of 4 women with polycystic ovary syndrome meet diagnostic criteria for binge eating disorder report of binge-purge eating patterns is also common. The exact mechanism or mechanisms linking bulimia nervosa and binge eating disorder with polycystic ovaries remain to be clarified.
Menstrual disturbance: a common presenting symptom
Endocrine and physiological abnormalities including primary and secondary amenorrhea and menstrual dysfunction are commonly associated with eating disorder pathophysiology. Although no longer a requirement for diagnosis, approximately 90% of women with anorexia nervosa report amenorrhea, the majority of which report secondary amenorrhea. Although there is a clear association between reduction of body weight, nutritional status, and amenorrhea, some women with low body weight have normal or infrequent menses.
Amenorrhea also precedes weight loss in approximately 20% of women with anorexia nervosa. Importantly, nutritional status, not psychiatric status, seems to be the determining factor that differentiates between women with and without amenorrhea. Studies report few meaningful differences in the course of illness, psychiatric comorbidity, eating disorder-related cognitions, or treatment outcome between menstruating and nonmenstruating patients. Thus, the absence of a regular menstrual cycle can be thought of as one marker of nutritional insufficiency but not pathognomonic for the disease.
Amenorrhea associated with anorexia nervosa is secondary to hypothalamic dysfunction, characterized by the suppression of gonadotrophin-releasing hormone pulsatility and subsequent estrogen deficiency. This malnutrition-induced functional hypothalamic amenorrhea is associated with a host of other physiological changes reviewed extensively elsewhere. These include a sick euthyroid pattern seen in chronic starvation (low endogenous levels of triiodothyronine, whereas reverse triiodothyronine concentrations and thyroxine to triiodothyronine ratio values are elevated), overactivity of the hypothalamic–pituitary-adrenal axis (increased secretion of corticotrophin-releasing hormone, hypercortisolemia), and abnormalities in appetite-regulating hormones (hypoleptinemia, elevated peptide YY).
Hypoleptinemia signaling a state of energy deficiency may mediate the neuroendocrine changes seen in hypothalamic amenorrhea. Eumenorrheic women with anorexia nervosa have higher serum leptin levels and more body fat compared with ammenorheic women of similar weight, suggesting that fat mass may play a role in persevering menstrual function. Although leptin administration has been shown to normalize gonadotrophin-releasing hormone pulsatility and induce menstruation in normal-weight women, there is no established threshold by which leptin levels can predict return to or cessation of menstruation. The considerable overlap in serum leptin levels between menstruating and nonmenstruating women with anorexia nervosa makes this unlikely and underscores the notion that leptin is likely one of many factors implicated in the pathophysiology of hypothalamic amenorrhea in women with anorexia nervosa.
Although most frequently recognized in anorexia nervosa, menstrual dysfunction is a frequent presenting symptom for women across the eating disorder diagnostic spectrum including those with bulimia nervosa, binge eating disorder, and other specified feeding or eating disorders. In a study of 1705 women with eating disorders, approximately 40% of women with bulimia nervosa presented with amenorrhea and 19% reported oligomenorrhea.
Menstrual irregularities were also common (49%) in those with other specified presentations. Specific eating disorder symptoms show a distinct relationship to patterns of menstrual dysfunction. Although severe caloric restriction, excessive exercise, and low lifetime body mass index are commonly associated with the cessation of menses, binge eating and/or purging are associated with irregular or normal menstruation. Binge eating may provide sufficient, albeit irregular, caloric absorption to preserve the menstrual cycle to some extent. Conversely, a high frequency of vomiting may interfere with the menstrual cycle by increasing dopaminergic and opioid activity. Moreover, there is a noteworthy association between bulimia nervosa, binge eating disorder, and polycystic ovary syndrome, which itself is marked by menstrual irregularities and disruption.
Large fluctuations and dietary intake brought on by binging, restriction, and/or purging may facilitate the expression of morphologically polycystic ovaries, potentially mediated by abnormalities in insulin secretion. Ultrasonography has revealed abnormal ovarian morphology in bulimia nervosa, with greater than 75% of patients having polycystic ovaries. In addition, 1 of 4 women with polycystic ovary syndrome meet diagnostic criteria for binge eating disorder report of binge-purge eating patterns is also common. The exact mechanism or mechanisms linking bulimia nervosa and binge eating disorder with polycystic ovaries remain to be clarified.
Skeletal and reproductive consequences
Amenorrhea stemming from low body weight has a devastating effect on bone mineral density and has serious long-term consequences. Eighty-five percent of women with anorexia nervosa have bone mineral density values more than 1 SD below healthy age-matched peers.
Early intervention is key, particularly given the frequent onset of anorexia nervosa during adolescence, which is a critical period for optimizing bone health and achieving peak bone mass. Failure to intervene early places patients at risk for potentially irreversible skeletal damage. Women who are amenorrheic and are of low weight have an annual rate of decline in bone mineral density of –2.6% at the spine and –2.4% at the hip. Risk of fracture is increased 7-fold.
Complications are long term with population-based retrospective cohort studies demonstrating the cumulative incidence of fracture to be 57%, with most complications occurring 2 decades after diagnosis. Even though women with bulimia nervosa may also present with low bone density, research suggests that such an association is largely accounted for by a history of anorexia nervosa.
The high prevalence of absent or infrequent menstruation observed in eating disorders has led many to hypothesize that women with eating disorders would be unlikely to conceive. On the one hand, there is some research to support this notion. A study of 2257 women treated for anorexia nervosa at a hospital eating disorder clinic reported compromised long-term reproductive health, with a pregnancy rate for women with histories of anorexia nervosa less than half the rate of matched controls. Research also consistently highlights a complex interaction between bulimia nervosa/binge eating disorder, frequent comorbid conditions of polycystic ovary syndrome and obesity, and infertility. Prevalence of current or past eating disorders in patients presenting to fertility clinics is also significantly higher than the general population.
On the other hand, 2 large population-based cohort studies suggest that a history of anorexia nervosa does not have long-term effects on fertility. Rates of pregnancy, reported infertility, and infertility treatment in these studies were not significantly different in women with histories of anorexia nervosa compared with the normal population. Similarly, Crow et al found that despite high rates of menstrual dysfunction, bulimia nervosa does not seem to have an impact on a later ability to conceive.
Although research is mixed regarding the impact on long-term fertility, a heightened risk of unplanned pregnancy is clear. Despite the high prevalence of menstrual disturbance in women with anorexia nervosa, the risk of an unplanned pregnancy is 2-fold in anorexic women compared with non–eating-disordered women. Women with bulimia nervosa also report a high rate of unplanned pregnancy and a greater likelihood of conceiving with oligomenorrheic menstrual status compared with quiescent bulimia nervosa controls.
As suggested by Bulik et al, irregular or absent menstruation coupled with the belief that menstrual irregularities reduce the risk of conception may be related to reduced adherence to contraception guidelines. Impulsive personality traits including promiscuity and unprotected sex are often characteristics of bulimic spectrum disorders and may also contribute to an increased risk of unplanned pregnancy. Thus, it is clear that women can and do become pregnant despite the neuroendocrine sequelae of eating disorders.
Bridging menstrual status assessment with eating disorder screening
Although menstrual dysfunction is certainly not unique to or always indicative of an eating disorder, it is one symptom that indicates a need for further inquiry and can provide a starting point for discussion of dietary and weight-control habits. Indeed, the American College of Obstetricians and Gynecologists has advocated that menstrual status be considered a “vital sign” at routine clinic visits, given the potential long-term implications of menstrual dysfunction.
Clinical assessment should include a thorough assessment of menstrual history including, age at first menses, regularity of menses, length of flow, and number of menstrual cycles in the preceding year. However, it is also important to remember that many women with eating disorders have normal menstrual cycles and that the assessment of menstrual status may be complicated by other factors such as age, hormonal contraceptives, or the recent initiation of antidepressant treatment. Indeed, universal screening is critical, given the long-term, devastating impact that eating disorders have on physical and emotional health. Discussion of menstrual health is only an entry point for a discussion. Inquiry into eating and weight behaviors should occur, regardless of patient endorsement or denial of menstrual dysfunction.
Several measures have been developed to assist primary care providers in screening for eating disorders. Perhaps the most widely adopted measure, the SCOFF, is a quick 5 item screening tool for eating disorders that has been validated in a number of primary care–based studies. Although this measure demonstrates high specificity, it should be used with caution because more recent studies have found low levels of sensitivity, particularly in diverse samples.
Eating disorders occur across the age spectrum and in individuals of all racial/ethnic backgrounds. Recognition of disordered eating patterns in racial/ethnic minorities is low, despite evidence that minorities are equally or more likely to suffer from eating disorders, particularly bulimia nervosa and binge eating disorder. This, coupled with recent revisions to the criteria such that binge eating disorder is no longer a provisional diagnosis, suggests that a substantial number of individuals with eating disorders may be missed. Thus, providers should not solely rely on the SCOFF and must directly ask all patients about their exercise and dietary habits, including a history of binging and purging, weight loss or gain, and recent psychosocial stressors.
The Bulimic Investigatory Test, Edinburgh and the Eating Disorder Examination-Questionnaire are 2 examples of reliable and valid self-report measures that can be given in the waiting room prior to the appointment and quickly reviewed by a clinician to provide another starting point for in-person assessment. Because early diagnosis and multidisciplinary treatment results in better outcomes, it is important to recognize that the report of any disordered eating behavior requires treatment, even if a woman denies or minimizes its significance. Open and ongoing communication is key to the identification and prevention of possible gynecological or obstetric complications.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
