Chapter 13 Cynthia M. Bulik, Sara E. Trace, Susan C. Kleiman, and Suzanne E. Mazzeo Eating disorders represent a category of partially overlapping syndromes, all of which have some clinical features marked by eating dysregulation. We will focus our discussion on anorexia nervosa (AN), bulimia nervosa (BN), and binge eating disorder (BED), which represent the primary eating disorders listed in DSM-5. Feeding disorders, such as pica, rumination disorder, and avoidant/restrictive food intake disorder—all more common in, but not exclusive to children—will not be covered here. Eating disorders are serious mental illnesses that are influenced by both genetic and environmental factors. The syndromes are partially overlapping, as considerable diagnostic flux occurs over time, with individuals migrating from one clinical presentation to another, and because several diagnostic features are shared across disorders. Nonetheless, pure forms of each of the presentations also exist. AN, the most visible eating disorder, is a serious psychiatric illness characterized by an inability to maintain a normal healthy body weight or, in individuals who are still growing, failure to make expected increases in weight (and often height) and bone density. Despite increasing weight loss and frank emaciation, individuals with AN strive for additional weight loss, see themselves as fat even when they are severely underweight, and often engage in unhealthy weight-loss behaviors (e.g., purging, dieting, excessive exercise, and fasting). AN is characterized by low weight; however, how one defines low weight is somewhat complicated. DSM-5 highlights restriction of energy intake relative to requirements, leading to a significantly low body weight, and embeds that in the context of the individual’s age, sex, developmental trajectory, and physical health. Even when at low weight, people with AN experience an intense fear of gaining weight or of becoming fat, or they engage in persistent behavior that interferes with weight gain. The behavior and cognitions of individuals with AN vigorously defend low body weight. Other aspects of the diagnostic criteria include a three-part criterion of which only one component is necessary: disturbance in the way in which one’s body weight or shape is experienced, undue influence of body weight or shape on self-evaluation, or persistent lack of recognition of the seriousness of the current low body weight. In the past, amenorrhea of 3 months or longer duration was a diagnostic criterion for AN. Wisely, this has been eliminated, as there are no meaningful differences between individuals with AN who do and do not menstruate (Gendall et al., 2006; Watson & Andersen, 2003). Although not diagnostic, cessation of menstruation can be a useful indicator of severity and resumption of menses a factor in determining recovery. AN presents either as the restricting subtype, in which low weight is achieved and maintained through energy restriction and increased physical activity only, or as the binge-eating/purging subtype, in which the individual has been regularly engaging in binge eating or purging behavior (i.e., self-induced vomiting or the misuse of laxatives, diuretics, or enemas) over the past 3 months. BN is characterized by recurrent binge eating episodes, defined as eating an unusually large amount of food in a short period of time (∼2 hours) while experiencing a sense of loss of control over the eating episode. In addition, bulimia includes recurrent inappropriate compensatory behaviors (e.g., self-induced vomiting, laxative, diuretic, or other medication misuse, fasting, or excessive exercise). In individuals with BN, self-evaluation is unduly influenced by body shape and weight. Binge eating and compensatory episodes occur on average once a week for at least 3 months. BN is only diagnosed if AN criteria are not met. Thus, to be diagnosed with BN, individuals should have a body mass index (BMI) greater than 18.5 kg/m2 in adults (i.e., the lower bound of normal weight according to the World Health Organization [1992] and the Centers for Disease Control). BN onset most frequently occurs in adolescence or early adulthood, although it can occur at any point across the life span (American Psychiatric Association [APA], 2013). BN can occur at any body weight (with the exception of the requirement to diagnose AN binge-eating/purging type if criteria for AN are met). BN tends to be overrepresented in women; however, it has been argued that BN diagnostic criteria are gender-biased, leading to underdetection in men. Men who seek treatment for BN tend to manifest a greater reliance on nonpurging forms of compensatory behavior, such as excessive exercise (Anderson & Bulik, 2003; Lewinsohn, Seeley, Moerk, & Striegel-Moore, 2002). It is important to consider such gender differences in the clinical presentation of BN to revise prevalence estimates of this diagnosis (Anderson & Bulik, 2003). In DSM-5, BED finally received recognition as a stand-alone disorder after years of being categorized as a disorder “worthy of further study.” Binge eating was first noted in a subset of obese individuals by Stunkard in 1959 (Stunkard, 1959). BED has had a slow and controversial evolution in the psychiatric nosology for eating disorders (Fairburn, Welch, & Hay, 1993; Spitzer et al., 1993; Walsh, 1992). BED is marked by recurrent binge eating (at least weekly for 3 months, as in BN) and a sense of lack of control over eating during the episode, but in the absence of regular compensatory behaviors. Unlike BN, the diagnostic criteria for BED include descriptions of the binge experience. To meet criteria, an individual must experience distress regarding the binge eating and at least three of the following: eating much more rapidly than normal, eating until feeling uncomfortably full, eating large amounts of food when not feeling physically hungry, eating alone because of feeling embarrassed by how much one is eating, or feeling disgusted with oneself, depressed, or very guilty afterward. It remains a curiosity why these descriptors remained in the BED criteria when they are not in the BN criteria; presumably, this was related to ensuring that individuals who simply overeat were not misdiagnosed as having BED. BED can occur at any body weight and is only diagnosed if neither AN nor BN criteria are met. Other Specified Feeding or Eating Disorder (OSFED) is a new category in DSM-5, which replaces the historical Eating Disorder Not Otherwise Specified (EDNOS). The reorganization in DSM-5 occurred in part because BED became a stand-alone diagnosis and in part because, historically, far too many individuals with eating disorders received a diagnosis of EDNOS, rendering it the most frequently diagnosed eating disorder. This alerted many researchers and clinicians to the fact that the diagnostic system was in need of revision so that a greater number of individuals could be captured under the hallmark categories of AN, BN, and BED. OSFED applies to presentations with symptoms characteristic of a feeding and eating disorder but to presentation where full diagnostic criteria are not met. Given that this is indeed a new category, we have very little epidemiologic data that reflect the new diagnosis. Research over the next several years will reveal how the transformation of the diagnostic schema has reshuffled the prevalences of the various disorders. OSFED includes a useful category of atypical AN, in which an individual meets all criteria for AN except that their weight falls within or above the normal weight range. This would capture, for example, an individual who was obese who precipitously lost a large amount of weight and exhibited all of the psychological features of AN, but because of the weight at which the weight loss started, still fell in the normal weight range. Other presentations under OSFED include BN and BED of low frequency or limited duration, purging disorder (which is purging behavior in the absence of binge eating), and night eating syndrome, in which individuals report recurrent episodes of night eating, marked by eating after awakening from sleep or by excessive food consumption after the evening meal. Based on previous research with EDNOS, what we do not expect to change is that being in the OSFED category in no way implies that an individual has a less serious disorder. The severity of pathology and psychosocial impairment is comparable among individuals with EDNOS, AN, and BN (Fairburn & Bohn, 2005; Keel, Gravener, Joiner, & Haedt, 2010). Clinical descriptions of EDNOS are consistent in stating that most cases have features similar to AN and BN (Crow, Agras, Halmi, Mitchell, & Kraemer, 2002; Waller, 1993; Walsh & Garner, 1997). Three studies (Fairburn & Cooper, 2007; Ricca et al., 2001; Turner & Bryant-Waugh, 2003) using the Eating Disorder Examination (EDE; Cooper & Fairburn, 1987) found that individuals with EDNOS presented with significant cognitive symptomatology related to eating, shape, and weight, suggesting that these syndromes are clinically significant. With the publication of DSM-5 in 2013, investigation of the validity of the new classification system is an important research focus. Some advances of the DSM-5 system include attention to stages of illness. In the past, for example, if someone had met criteria for AN and then began to recover, she or he might have received a new diagnosis of EDNOS. In DSM-5, there is now the option to include the specifier of “in partial remission” if, after having met full criteria, weight has normalized but the psychological features remain, and “full remission” if, after having met full criteria, no criteria have been met for a sustained period of time. In addition, severity specifiers also exist, and are currently based on body mass index (BMI), with mild AN being ≥ 17 kg/m2, moderate 16–16.99 kg/m2, severe 15–15.99 kg/m2, and extreme <15 kg/m2. In addition to the core diagnostic features, individuals with AN often manifest a specific cluster of personality traits, including perfectionism, obsessionality, anxiety, harm avoidance, and low self-esteem (Cassin & von Ranson, 2005; Fassino, Amianto, Gramaglia, Facchini, & Abbate Daga, 2004; Klump et al., 2000). Furthermore, both these personality characteristics and anxiety disorders often precede AN onset (Bulik, Sullivan, Fear, & Joyce, 1997; Kaye et al., 2004). Major depression and anxiety disorders frequently co-occur with AN (Bulik et al., 1997; Fernandez-Aranda et al., 2007; Godart, Flament, Perdereau, & Jeammet, 2002; Godart, Flament, Lecrubier, & Jeammet, 2000; Kaye et al., 2004), and longitudinal research suggests that depression often persists following recovery from AN (Sullivan, Bulik, Fear, & Pickering, 1998). Some personality features common among individuals with AN are also manifested by many women with BN, such as high harm avoidance, perfectionism, and low self-esteem. However, other personality features appear more specific to BN, including elevated novelty seeking and impulsivity, low self-directedness, and low cooperativeness (Bulik, Sullivan, Carter, & Joyce, 1995; Fassino et al., 2004; Steiger et al., 2004). Further refinements of the components of impulsivity suggest that negative urgency, or the tendency to act rashly when distressed, is the facet of impulsivity most strongly associated with bulimia (Fischer, Smith, & Cyders, 2008). Comorbid psychiatric disorders are very common among individuals with BN, occurring among nearly 80% of patients (Fichter & Quadflieg, 1997). These comorbidities include anxiety disorders, major depression, dysthymia, substance use, and personality disorders (Braun, Sunday, & Halmi, 1994; Brewerton et al., 1995; Bulik et al., 2004; Perez, Joiner, & Lewinsohn, 2004). Finally, BED also commonly co-occurs with numerous other psychiatric diagnoses, including mood, anxiety, and substance abuse disorders (Grucza & Beirut, 2007; Johnson, Spitzer, & Williams, 2001; Marcus, Wing, & Fairburn, 1995; Striegel-Moore et al., 2001; Wilfley, Freidman, et al., 2000). Data from the National Comorbidity Survey Replication (Hudson, Hiripi, Pope, & Kessler, 2007) indicate that BED is a chronic condition associated with significant impairment in daily functioning. Global data from the World Heath Organization World Mental Health Surveys indicate that BED and BN are associated with significantly increased education in women. Early-onset BED predicted reduced odds of marriage in women and reduced odds of employment in men, while early-onset BN predicted increased odds of current work disability in both sexes. Both BED and BN were associated with significantly increased days of role impairment, although much of the role impairment was accounted for by the presence of comorbid disorders (Kessler, Shahly, et al., 2013). Finally, those individuals with BED who are overweight or obese are at risk for medical complications (Hudson et al., 2007). Yet, the negative psychological impact of BED does not appear to be attributable to obesity. Obese individuals with BED report substantially poorer psychological functioning than do obese individuals without BED (Grucza, Przybeck, & Cloninger, 2007), and normal weight and overweight individuals with BED report equivalent psychological features of disordered eating and depression (Dingemans & van Furth, 2012). Available epidemiologic data on eating disorders reflect DSM-IV diagnostic criteria, because sufficient time has not yet elapsed for studies to be conducted on the new classifications. Lifetime prevalence estimates of DSM-IV AN, BN, and BED from a nationally representative population sample over age 18 are 0.9%, 1.5%, and 3.5% in women, and 0.3%, 0.5%, and 2.0% in men, respectively (Hudson et al., 2007). The prevalence of subthreshold AN, defined as at least one criterion short of threshold, is greater and ranges from 0.37% to 1.3% (Hoek, 1991). The gender ratio for AN is approximately 9:1, women to men (APA, 1994). Awareness of these disorders has increased; however, the data on changing incidence are conflicting. Some studies report increasing incidence of AN or increases in disordered eating behavior (such as strict dieting or fasting for weight or shape control) that are associated with AN (e.g., Hay, Mond, Buttner, & Darby, 2008; Lucas, Crowson, O’Fallon, & Melton, 1999; Eagles, Johnston, Hunter, Lobban, & Millar, 1995; Jones, Fox, Babigan, & Hutton, 1980; Møller-Madsen & Nystrup, 1992), whereas others describe stable prevalence (e.g., Smink, van Hoeken, & Hoek, 2012; Hoek, 2006; Currin, Schmidt, Treasure, & Jick, 2005; Pawluck & Gorey, 1998; Hall & Hay, 1991; Hoek et al., 1995). The peak age of onset for AN is between 15 and 19 years (Lucas, Beard, O’Fallon, & Kurland, 1988), an age group for which incidence has been increasing (Smink et al., 2012). However, reports suggest new-onset cases in mid- and late life (Gagne et al., 2012; Mangweth-Matzek et al., 2006; Beck, Casper, & Andersen, 1996; Inagaki et al., 2002) and increasing presentations in children (Rosen, 2010). The prevalence of BN in the United States is estimated to be 1.5% for women and 0.5% for men (Hudson et al., 2007). The prevalence of subthreshold behaviors is considerably higher, with 4.9% of women and 4% of men endorsing any binge eating. Similar to AN, reports suggest that more children and older adults are presenting with BN (Marcus et al., 2007; Rosen, 2010). The prevalence of BED in the United States has been estimated at 3.5% for women and 2% for men (Hudson et al., 2007), while community surveys across 12 countries estimate the lifetime prevalence across both genders at 1.9% (Kessler, Shahly, et al., 2013). In a population-based study of female twins, 37% of obese women (BMI ≥ 30) reported binge eating (Bulik, Sullivan, & Kendler, 2002), representing 2.7% of the female population studied. Community studies of obese individuals have found a prevalence of BED between 5% and 8% (Bruce & Agras, 1992; Bruce & Wilfley, 1996). The sex distribution in BED is more equal than in AN or BN (Hudson et al., 2007) with few differences in prevalence across races or ethnic groups (Alegria et al., 2007; Marcus et al., 2007). Careful and accurate assessment of eating disorders, which are frequently complex and have multiple presentations, is critical for effective treatment and research. The general goal of psychological assessment is to elicit information that accurately describes symptomatology, accurately characterizes diagnostic profile, and indicates appropriate treatment recommendations (Peterson & Mitchell, 2005). Assessing individuals with eating disorders is often challenging secondary to denial of the illness and hidden signs and symptoms (Palmer, 2003; Schacter, 1999; Tury, Gulec, & Kohls, 2010; Vitousek, Daly, & Heiser, 1991). The use of active listening skills is important for developing rapport (Keel, 2001), and motivational interviewing techniques (Miller & Rollnick, 2002), which encourage rolling with resistance, avoiding arguments, and expressing empathy, are often helpful for conducting a successful assessment. Clinical interviews in eating disorders are used to elicit the patient’s perspective of the development of his or her difficulties and frequently include the reason for the assessment/primary complaint, history of present illness, medical complications, treatment history, and coexisting conditions (Peterson, 2005). A combination of structured interviews, self-report measures, and medical assessments may also be employed to obtain a more complete clinical picture. In the case of minors, corroborating information, such as reports from parents or school officials, is additionally informative (Lock, Le Grange, Agras, & Dare, 2001). Structured interviews are essential for clarifying differential diagnostic issues and assessing psychiatric comorbidity. Structured interviews are advantageous in that they allow for active involvement of the interviewer, who can help clarify concepts or answer questions that may arise during the assessment. Obvious drawbacks to structured interviews include greater financial cost and clinician burden (Grilo, 2005). For untrained interviewers, the two dominant instruments for assessing Axis I pathology are the Diagnostic Interview Schedule (DIS; Robins, Helzer, Croughan, & Ratcliff, 1981) and the Composite International Diagnostic Interview (CIDI; World Health Organization, 1990). The various versions of the Structured Clinical Interview for DSM-IV (SCID; First, Spitzer, Gibbon, & Williams, 2002), which has excellent validity and reliability (Grilo, 2005; Zanarini et al., 2000), are recommended for assessing Axis I pathology in adults by trained interviewers. Several clinician-based structured or semistructured interviews have been developed specifically for assessing eating disorder symptomatology. The Eating Disorder Examination (EDE; Cooper & Fairburn, 1987) is well-established (Wilfley, Schwartz, Spurrell, & Fairburn, 2000) and widely used. It includes 33 items that measure behavioral and psychological traits in AN and BN and, with the exception of the diagnostic items, focuses on the 28 days preceding the assessment. Items are rated on a 7-point Likert-type scale, with higher scores indicating greater pathology, and comprise the following scales: dietary restraint, eating concern, weight concern, and shape concern. The EDE has high interrater reliability (Cooper & Fairburn, 1987; Grilo, Masheb, Lozano-Blanco, & Barry, 2004; Rizvi, Peterson, Crow, & Agras, 2000), adequate internal consistency (Beumont, Kopec-Schrader, Talbot, & Touyz, 1993; Cooper, Cooper, & Fairburn, 1989), and good discriminative validity for distinguishing those with eating disorders from healthy individuals (Cooper et al., 1989; Wilson & Smith, 1989). Other popular structured interviews for assessing disordered eating include the Interview for Diagnosis of Eating Disorders (IDED; Williamson, 1990) and the Structured Interview for Anorexic and Bulimic Disorders (SIAB-EX; Fichter et al., 1998). For a full review of these and other structured interviews in eating disorders, see Grilo, 2005. The IDED-IV (Kutlesic, Williamson, Gleaves, Barbin, & Murphy-Eberenz, 1998) is another semistructured interview primarily used for differential diagnosis of DSM-IV AN, BN, and EDNOS. The IDED-IV differs from the EDE in that it does not focus on frequency and severity data, but rather on differential diagnosis. Four studies support the psychometric properties of this instrument (Kutlesic et al., 1998). The current version of the SIAB-EX (Fichter et al., 1998) assesses specific criteria for AN and BN (including subtypes), consistent with both the DSM-IV and the ICD-10. There is also an algorithm that allows the data to be used to generate the BED research diagnosis and other eating disorder syndromes under the EDNOS category. The SIAB-EX has demonstrated good internal consistency, factor structure, interrater reliability, and convergent and discriminant construct validity (Fichter & Quadflieg, 2000, 2001). Overall, the EDE and the SIAB-EX have been shown to produce generally similar findings. However, areas of divergence do exist, many of which could be attributable to the differences in criteria and time frames for assessment (Fichter & Quadflieg, 2001). Many self-report measures are available for assessing disordered eating both in research and clinical settings. Self-report assessments can be used for a variety of purposes, including identifying clinical features, quantifying symptoms, and verifying diagnoses. They are particularly useful for assessing change over time and are time and cost effective because they can be completed independently by the patient (Peterson & Mitchell, 2005). Two of the most widely used self-report questionnaires for assessing disordered eating include the Eating Disorder Inventory (EDI) and the Eating Disorder Examination–Questionnaire (EDE-Q). The EDI (Garner, Olmsted, & Polivy, 1983, 1984), which assesses eating disorder symptoms and associated psychological traits, is useful for differentiating levels of eating disorder severity and for assessing treatment outcome (Williamson, Anderson, Jackman, & Jackson, 1995). This assessment is described by the authors as “investigator-based,” emphasizing that it is the investigator’s job to make final judgments about what symptoms and behaviors are present (e.g., to determine what constitutes a binge). The EDI has 64 questions answered on a 6-point Likert-type scale and comprises the following eight subscales: drive for thinness, bulimia, body dissatisfaction, ineffectiveness, perfectionism, interpersonal distress, interoceptive awareness, and maturity fears. A revised version of the EDI, the EDI-2, was published in 1991 and includes 27 additional questions. The eight scales from the EDI were retained, and three additional scales—asceticism, impulse regulation, and social insecurity—were incorporated (Garner, 1991). The third version of the scale, EDI-3 (Garner, 2004), retained the same items as the EDI-2 but has a slightly different factor structure (Garner, Olmsted, & Polivy, 2008). It contains 91 items rated on a 0–4 point scoring system. The three subscales assessing eating pathology added in the EDI-2 (drive for thinness, bulimia, and body dissatisfaction) remain largely unchanged, and the general psychology subscales include low self-esteem, personal alienation, interpersonal insecurity, interpersonal alienation, interoceptive deficits, emotional dysregulation, perfectionism, asceticism, and maturity fears. Scoring for the EDI-3 includes the following six composite scores: (1) eating disorder risk, (2) ineffectiveness, (3) interpersonal problems, (4) affective problems, (5) over control, and (6) general psychological maladjustment, as well as infrequency and negative impression scores. The EDI-3 has yielded reliable and valid scores (Garner, 2004). The EDE-Q (Fairburn & Beglin, 1994), another widely used self-report measure of eating disorder symptoms, assesses severity of eating pathology and associated disturbances over the past 28 days. It is most often used in research, but it can be applied in clinical settings as well (Peterson & Mitchell, 2005). The EDE-Q was adapted from the structured interview EDE (Cooper & Fairburn, 1987), and like the EDE, it consists of 33 items and four subscales (restraint, eating concern, shape concern, and weight concern). The subscales and total scores are based on averages from 0 to 6, with higher scores indicating greater pathology. The EDE-Q has been described as an accurate method for assessing binge eating (Wilson, Nonas, & Rosenbaum, 1993) and shows acceptable reliability and validity (Fairburn & Cooper, 1993). There are numerous other self-report assessments for eating disorders, including the Multiaxial Assessment of Eating Disorder Symptoms (MAEDS; Anderson, Williamson, Duchmann, Gleaves, & Barbin, 1999), the Stirling Eating Disorder Scales (SEDS; Williams et al., 1994), the Anorexia Nervosa Inventory for Self-Rating (ANIS; Fichter & Keeser, 1980), the Three Factor Eating Questionnaire (TFEQ; Stunkard & Messick, 1985), the Binge Eating Scale (BES; Gormally, Black, Daston, & Rardin, 1982), and the Questionnaire for Eating and Weight Patterns-Revised (QEWP-R; Yanovski, 1993). A full review of these and other self-report measures for assessing disordered eating can be found in Peterson and Mitchell (2005) or Tury, Gulec, and Kohls (2010). Careful medical assessment, both initially and as indicated throughout the duration of eating disorder treatment, is critical for effective treatment (Crow, 2005). It is also important for Emergency Medicine physicians to be able to screen for and recognize patients with eating disorders, and to be aware of their medical complications and psychiatric comorbidities, in order to carry out a successful therapeutic intervention (Trent, Moreira, Colwell, & Mehler, 2013; Mascolo, Trent, Colwell, & Mehler, 2012). Documentation of medical complications is imperative, not only for treatment planning but also for service authorization by insurance companies. Although all eating-disorder presentations require medical monitoring, low-weight patients, individuals with purging behaviors, and obese individuals with binge-eating behavior (or a combination of these behaviors) are typically at the greatest risk for medical complications (e.g., Crow, Salisbury, Crosby, & Mitchell, 1997; Harris & Barraclough, 1998; Kohn, Golden, & Shenker, 1998). Low-weight individuals are particularly vulnerable to medical morbidity and mortality (Harris & Barraclough, 1998). A BMI below 13 is associated with less favorable outcome (Hebebrand et al., 1997), and low weight is associated with increased likelihood of sudden cardiac death. AN, BN, and EDNOS are all associated with increased mortality (Crow et al., 2009). Evidence of medical complications might also encourage otherwise resistant patients to enter treatment. A standard initial assessment for low-weight individuals should include a complete blood count, an electrolyte battery (including phosphorus, calcium, and magnesium), an electrocardiogram, liver function tests, and a dual-energy X-ray absorptiometry (DEXA) scan (Crow, 2005). Blood pressure and pulse should also be documented, as dehydration can lead to orthostatic hypotension. The patient should be monitored carefully through the re-feeding process, because provision of adequate calories may lead to a drop in serum phosphorus, which is associated with mortality (Kohn et al., 1998) both in hospital (Ornstein, Golden, Jacobson, & Shenker, 2003) and outpatient settings (Winston & Wells, 2002). Electrolyte disturbance is the most commonly recognized complication of purging behaviors (Crow et al., 1997). Although not sensitive to vomiting frequency, hypokalemia is a marker of vomiting behavior (Crow et al., 1997). Another common complication of self-induced vomiting is parotid hypertrophy, or painless swelling of the parotid glands, which may persist for months following cessation of purging (Ogren, Huerter, Pearson, Antonson, & Moore, 1987). Dental complications, including dental enamel erosion on the lingual surfaces of teeth (Little, 2002), may occur in individuals who vomit frequently and, thus, continued dental monitoring is important. A smaller number of individuals with purging behaviors report gastrointestinal symptoms, including intestinal bleeding, hematemesis (vomiting blood), the passing of melanotic stools, or blood in the stools. Although rare, esophageal tears, gastric erosions, hemorrhoids, and gastric rupture may also occur (Cuellar, Kaye, Hsu, & Van Thiel, 1988; Cuellar & Van Thiel, 1986). Abuse of laxatives and emetics are also associated with significant medical morbidity. The use of syrup of Ipecac should signal a medical and cardiac evaluation, as it is associated with severe cardiac effects. BED, which is among the most common of eating disorder presentations, is often associated with co-occurring conditions (Crow, 2005), including Type II diabetes mellitus and obesity. There is some evidence to suggest that obese individuals with Type II diabetes mellitus who also binge eat experience worse outcomes than their non-binge-eating peers (Goodwin, Hoven, & Spitzer, 2003; Mannucci et al., 2002). Binge eating appears to be associated with medical problems independent of obesity (Bulik et al., 2002). Moreover, BED may confer a risk of developing metabolic syndrome (a cluster of related risk factors for atherosclerotic cardiovascular disease, including abdominal obesity, dyslipidemia, hypertension, and abnormal glucose metabolism) beyond the risk attributable to obesity alone (Hudson et al., 2010). It is critical to remember that not all individuals with BED are overweight or obese. We await further data on the health impact of BED in normal weight individuals. The growing interest in eating disorders over the past 20 years has resulted in the development of numerous assessment tools for research and clinical purposes. Accurate assessment of individuals with disordered eating requires a multidisciplinary approach to address both the psychological and biological factors underlying etiology. Although numerous psychological, social, and biological factors have been implicated as potential causes of eating disorders, few specific risk factors have been consistently identified across studies, and the etiology of these disorders is not fully understood (Jacobi, Hayward, de Zwaan, Kraemer, & Agras, 2004; Striegel-Moore & Bulik, 2007). Common risk factors across eating disorders include female sex, race, or ethnicity, childhood eating and gastrointestinal problems, elevated concerns about shape and weight, negative self-evaluation, prior history of sexual abuse and other adverse events, and presence of additional psychiatric diagnoses (Jacobi et al., 2004). Developmentally, prematurity, smallness for gestational age, and cephalohematoma have been identified as possible risk factors for AN (Cnattingius, Hultman, Dahl, & Sparen, 1999). Current studies suggest that eating disorders are caused by a variety of factors, including both genetic (e.g., Trace, Baker, Peñas-Lledó, & Bulik, 2013; Bulik, Slof-Op’t Landt, van Furth, & Sullivan, 2007) and environmental influences (e.g., Becker & Hamburg, 1996; Garner & Garfinkel, 1980; Striegel-Moore & Bulik, 2007). Contemporary understanding of eating disorders incorporates both genetic and environmental factors into causal models. Previously, an overemphasis on sociocultural factors ignored the fact that, although social pressures toward thinness are ubiquitous, only a fraction of individuals exposed to these factors develop eating disorders. Therefore, a clearer understanding of vulnerability has led to the model that individuals who are more genetically predisposed to eating disorders are those who are also more vulnerable to environmental triggers of illness—typically ones that result in dieting, drive for thinness, and persistent negative energy balance. Environmental influences that might serve as eating disorder triggers include the media’s idealization of the thin body ideal and pressure to achieve an unrealistically thin body type (Irving, 1990; Levine & Harrison, 2004). Sociocultural models of disordered eating (Polivy & Herman, 1985; Striegel-Moore, Silberstein, & Rodin, 1986) suggest that the perception of a discrepancy between the self and the thin ideal leads to psychological discomfort. In turn, a desire to ameliorate this discomfort might result in eating disordered behavior. Striegel-Moore and Bulik (2007) report that cultural models of eating disorders are supported by the following: (a) the high percentage of female cases of disordered eating; (b) the increase in incidence of eating disorders in women coinciding with the decreasing body-weight ideal for women; (c) the reported higher incidence of eating disorders in cultures that emphasize thinness; and (d) the significant association between thin ideal internalization and disordered eating. In a community-based case-control study, Fairburn et al. (1998) found significant differences in exposure to risk factors between women with BED and healthy controls, but surprisingly few differences between women with BED and BN. Specifically, compared with controls, women with BED reported more adverse childhood experiences, parental depression, personal vulnerability to depression, and exposure to negative comments about weight, shape, and eating. Other studies have indicated that environmental factors, including parental and peer behaviors, contribute to both risk and protection from eating pathology (Enten & Golan, 2009; Twamley & Davis, 1999). For example, Twamley and Davis reported that low family pressures to control weight moderated the relation between exposure to thin norms and internalization of these messages. In addition, other environmental variables, including social pressure, could amplify or mitigate the risk of eating disorders (Striegel-Moore et al., 1986). For example, individuals exposed to peer teasing might be more likely to develop disordered eating (Thompson, Coovert, Richards, Johnson, & Cattarin, 1995; Thompson & Heinberg, 1993). Similarly, individuals from higher social classes might be more prone to develop disordered eating, as they presumably have more time, attention, and resources available to focus on the achievement of cultural beauty ideals (Striegel-Moore & Bulik, 2007). Although these factors might influence eating disorder etiology, they are likely not solely responsible for their development (Striegel-Moore & Bulik, 2007). Personality traits such as perfectionism, as well as social anxiety, elevated weight, and high impulsivity, might also play important etiological roles. These sociocultural and environmental factors likely combine with genetic influences (Strober, Freeman, Lampert, Diamond, & Kaye, 2000) to contribute to the development of disordered eating, as is described in the next section. The conceptualization of eating disorders has evolved rather radically across time (Vemuri & Steiner, 2007). Previously dominant sociocutural and psychodynamic theories have been supplanted by a biopsychosocial model. This evolution can be attributed in part to a systematic series of family twin and molecular genetics investigations of eating disorders, which have supported the role of familial and genetic factors in liability to eating disorders (Bulik et al., 2006; Klump, Miller, Keel, McGue, & Iacono, 2001). In this section, we review results of family, twin, and molecular genetic studies (for a more thorough review see Trace et al., 2013). Family studies investigate the degree to which a particular trait runs in families. Although they are a valuable tool, family studies cannot tell us why a trait runs in families—whether due to genetic factors, environmental factors, or some combination of both. The familial nature of AN is well-established. For example, first-degree relatives of patients with AN (parents, children, and siblings) are 11 times more likely to have AN during their lifetime than first-degree relatives of individuals who have never had AN (Strober et al., 2000). Population-based twin studies have provided additional support for the familiality of AN. Twin studies allow us to examine familial components of disordered eating by comparing similarities and differences in eating problems between monozygotic twins (MZ) and dizygotic twins (DZ). MZ twins are generally assumed to share 100% of their genetic material, whereas DZ twins, on average, share 50% of their genetic material (like brothers and sisters). Variance in liability to a disorder can be dissected into additive genetic factors, shared environmental factors, and unique environmental factors. Additive genetic factors refer to the cumulative effects of many genes, each of which makes a small to moderate contribution. Shared environmental factors reflect environmental influences that affect both members of a twin pair and are believed to make twins more similar. Unique environmental factors (including measurement error), on the other hand, reflect environmental factors that only one twin is exposed to. Unique environmental factors are believed to make twins dissimilar. Twin studies have yielded heritability estimates between 28% and 74% for AN, with the remaining variability largely attributed to unique environmental factors (Klump et al., 2001; Kortegaard et al., 2001; Bulik et al., 2006). Although twin studies can reveal the proportion of individual differences in a disorder that are due to genetic factors, they are unable to identify which specific genes are involved. Molecular genetic studies provide greater clarity regarding which genes influence risk for a trait or disorder. Association studies examine a genetic variant’s association with a trait; if the variant and trait are correlated, there is said to be an association between the two. Association studies that involve a single gene or set of genes that have a hypothesized association with the trait under study are referred to as candidate gene studies. Molecular genetic designs that do not focus on one particular gene or set of genes include linkage and genome-wide association studies (GWAS). Linkage identifies chromosomal regions that house predisposing or protective genes and allow us to narrow the search from the entire human genome to specific regions. GWAS examines 300,000 to 1,000,000 genetic markers scattered across the genome, comparing cases with the trait to controls. If a genetic variant is more frequent in cases, the variant is said to be associated with the trait. GWAS represents an agnostic search of the human genome and as such is a genetic discovery tool. Decades of candidate gene association studies for AN have examined primarily genes involved in the serotonergic, catecholaminergic, and dopaminergic systems and those affecting appetite and weight regulation. The practice of preselecting a single gene based on presumed biological involvement has fallen out of favor, and it has given way to genome-wide approaches (described next). Historically, using candidate gene approaches, the serotonergic system received significant attention, and results regarding its importance to eating disorders are inconclusive. One meta-analysis of studies investigating 5-HTTLPR and AN suggests that carriers of the short allele are at increased risk for this eating disorder (Calati, De Ronchi, Bellini, & Serretti, 2011). A comprehensive review of all candidate gene association studies conducted for AN (175 association studies of 128 polymorphisms related to 43 genes) points to promising although not conclusive evidence for genes related to mood regulation [brain-derived neurotrophic factor (BDNF) and SK3 channel], the hedonic reward system [catecholamine-O-methyltransferase (COMT) and opioid receptor-1 (OPRD1)], and appetite [agouti-related protein (AGRP)] (Rask-Andersen, Olszewski, Levine, & Schiöth, 2010). Linkage studies identified chromosomes 1, 4, 11, 13, and 15 as possible regions of interest in AN (Bacanu et al., 2005; Devlin et al., 2002; Grice et al., 2002). A follow-up study of candidate genes on chromosome 1 revealed associations with the serotonergic (5-HTR1D) and opioidergic (OPRD1) neurotransmitter system (Bergen et al., 2003). Genome-wide approaches have been conducted. One Japanese study used deoxyribonucleic acid (DNA) pooling and included only 23K microsatellite markers (Nakabayashi et al., 2009); Wang et al. (2011), conducted GWAS in 1,033 female AN cases and 3,733 pediatric controls; however, no single nucleotide polymorphisms (SNPs), or DNA sequence variation, reached genome-wide significance, which is typical in samples this small. A GWAS conducted under the auspices of the Wellcome Trust Case Control Consortium 3, also underpowered, failed to identify SNPS that reached genome-wide significance (Boraska, in press). Large global efforts are underway to boost sample size in order to identify variants that influence risk for AN (Sullivan, Daly, & O’Donovan, 2012). Like AN, BN runs in families. First-degree relatives of individuals with BN are 4 to 10 times more likely to have the disorder themselves (Lilenfeld et al., 1998). In studies of female twins, the estimated heritability of BN ranges between 54% and 83% in females (see Slof-Op’t Landt et al., 2005, for a review). As is the case in AN, molecular genetic studies of BN have generally focused on the serotonergic, dopaminergic, catecholamineric, and appetite systems. Significant associations have emerged between BN and 5-HT2A and 5-HTTLPR. Several meta-analyses (Calati et al., 2011; Lee & Lin, 2010; Polsinelli, Levitan, & De Luca, 2012) have examined the association between 5-HTTLPR polymorphisms and BN, with the large majority suggesting no significant association between 5-HTTLPR polymorphisms and BN. Investigations exploring associations between other serotonin receptor genes and BN have yielded mixed results (see Scherag, Hebebrand, & Hinney, 2010, for a review). However, associations have been identified between several traits related to BN and the serotonin system, including minimum lifetime BMI (5-HT1B), impulsiveness (5-HT2A and 5-HTTLPR), and affective dysregulation in females (5-HTTLPR) (see Scherag et al., 2010, for a review). Furthermore, a gene–environment interaction was identified in one study; within a sample of individuals with BN, carriers of the 5-HTTLPR short allele who reported physical or sexual abuse also manifested greater sensation seeking, insecure attachment, and dissocial behavior (Steiger et al., 2007, 2008). The existence of this type of gene–environment interaction might explain some of the inconsistent results regarding serotonin to date. Studies investigating genes within the dopamine and catecholamine systems and those genes involved in appetite have also yielded inconsistent findings. Nisoli et al. (2007) examined the prevalence of TaqA1 polymorphisms of the DRD2 gene in individuals with eating disorders, including BN, and in controls. No significant associations were found between the A1+ allele in BN for either the A1/A1 or A1/A2 genotypes. Sporadic associations were found between BN and the dopamine transporter gene (DAT1) (Shinohara et al., 2004) and COMT (Mikolajczyk, Grzywacz, & Samochowiec, 2010), respectively. In addition, a few studies have identified an association between BN and preproghrelin (Miyasaka et al., 2006) and BDNF, yet these results require replication. Only one linkage study has been conducted for BN, which examined 308 multiplex families identified through a patient with BN. Significant linkage was found on chromosome 10 and another region on chromosome 14 met criteria for genomewide-suggestive linkage (Bulik et al., 2003). No GWAS of BN have been conducted to date. In sum, results of molecular genetic studies of BN remain inconclusive and are limited by the use of small samples, which provide relatively low power. The study of BED has burgeoned in the past decade. However, as the disorder has been more recently operationalized than AN and BN, less research on the genetics of BED has emerged. Nonetheless, extant family, twin, and molecular research largely suggests that familial and genetic factors influence risk for BED. A small number of family studies have been conducted (Fowler & Bulik, 1997; Hudson et al., 2006; Lee et al., 1999). With the exception of the Lee et al. investigation, these studies suggest that BED is familial. This has been further corroborated by twin studies. Two population-based twin studies have examined the heritability of BED (Javaras et al., 2008; Mitchell et al., 2010) and reported heritability estimates ranging from 39% to 45%. Candidate gene association studies of binge eating and BED have focused on neurotransmitter systems, such as the 5-HT and DA systems, and genetic variants implicated in appetite and obesity. One small case control investigation, comparing women with BED to normal women without BED, was conducted exploring the role of the 5-HTTLPR polymorphism in BED (Monteleone, Tortorella, Castaldo, & Maj, 2006). The homozygous long-allele and the heterozygous long-allele genotypes were found to be more prevalent in individuals with BED than those without BED. Results may suggest a role of the 5-HTTLPR polymorphism in BED; however, results should be considered preliminary, as the sample sizes in this study were small. Several investigations have also examined the role of DA polymorphisms, and particularly polymorphisms of the DRD2 gene, in BED (Davis et al., 2008; Davis et al., 2009; Davis et al., 2012). Overall, studies exploring the association of between BED and polymorphisms of the DRD2 gene have been inconsistent, likely due to small sample sizes and a lack of statistical power. The largest study to date (Davis et al., 2012) suggests a potential role of the DRD2 polymorphism Taq1A and C958T in BED; however, additional large-sample replication studies are needed. Genes associated with obesity have also been investigated for their potential role in BED, given the positive correlation between these conditions. MC4R (which is associated with obesity) was examined as an early candidate for BED (Branson et al., 2003), although this finding is not consistently replicated across studies (Hebebrand et al., 2004). Positive associations with 5-HTTLPR and DAT1, BDNF, and ghrelin have also been identified in BED (Davis et al., 2007; Monteleone, Tortorella, Castaldo, Di Filippo, & Maj, 2007; Monteleone, Tortorella, et al., 2006; Monteleone, Zanardini, et al., 2006; Shinohara et al., 2004); however, these results require confirmation and replication, and the field awaits more comprehensive genome-wide approaches. Neurobiological vulnerabilities contribute to eating disorder pathogenesis (Kaye, 2008; Kaye, Wierenga, Bailer, Simmons & Bischoff-Grethe, 2013; Treasure & Campbell, 1994), and brain structural and functional abnormalities are consistently found in individuals with eating disorders (Frank, Bailer, Henry, Wagner, & Kaye, 2004; Kaye, Fudge, & Paulus, 2009). In addition, numerous behavioral traits associated with AN, including premorbid anxiety, obsessive behaviors, negative emotionality, impaired cognitive flexibility, increased harm avoidance and perfectionism, and altered interoceptive awareness, are hypothesized to be related to underlying abnormalities or alterations in brain structure and function (Kaye et al., 2013). Marsh et al. (2011) reported evidence of deactivation in the inferior frontal gyrus and neural system encompassing the posterior cingulate cortex and superior frontal gyrus in female adolescents with bulimia in comparison to controls during the Simon spatial incompatibility task. This paradigm allowed them to observe abnormal patterns of frontostriatal activation in adolescents with bulimia when engaging in self-regulatory processes associated with conflict resolution. They suggested that this pattern could explain how feeding behaviors might be “released” from regulatory control in conflict situations, thereby perpetuating bulimic behaviors. Brain structural abnormalities in eating disorders have been investigated using computerized tomography (CT) and magnetic resonance imaging (MRI). Functional imaging studies, including positron emission tomography (PET), single photon emission computer tomography (SPECT), and functional magnetic resonance imaging (fMRI), have also been employed to provide information about the cerebral activity of a system or receptor being studied. Improvements in technology over the last decade, particularly in neuroimaging and genetics, have greatly enhanced our ability to characterize the complex neuronal systems involved in disordered eating (Kaye, 2008; Kaye et al., 2013). However, these techniques are still relatively new, and our understanding of the relation between biological vulnerabilities and subsequent changes in brain pathways contributing to disordered eating are limited. Neurobiological investigations of disordered eating are further complicated by state-related effects from changes in diet and weight, which impact neuronal processes. Brain imaging is not at a point yet where it can be used diagnostically; however, with the refinement of imaging hardware and improved models of the neurobiology and genetics of psychiatric illness, scientists are hopeful that in the future, imaging will allow us to untangle the complexities of eating disorders, predicting illness development, treatment response, and long-term prognosis (Frank, 2013). At present, central nervous system (CNS) dysregulation of neuropeptides (Bailer & Kaye, 2003) and monoamines (Bailer et al., 2007; Kaye, 2008) as well as brain structural abnormalities (Artmann, Grau, Adelmann, & Schleiffer, 1985; Heinz, Martinez, & Haenggeli, 1977; Joos et al., 2010; Krieg, Lauer, & Pirke, 1989), are implicated in the neurobiology of disordered eating. Neuropeptides involve a complicated interplay between the peripheral system and the CNS (Morton, Cummings, Baskin, Barsh, & Schwartz, 2006), and opioid peptides, corticotropin-releasing hormone (CRH), vasopressin, oxytocin, neuropeptide-Y, Peptide YY (PYY), cholecystokinin (CCK), leptin, ghrelin, and gastrin-releasing peptides are reported to play an important role in the regulation of feeding behavior (Bailer & Kaye, 2003; Monteleone, 2011; Yagi et al., 2012). A growing body of literature documents alterations in neuropeptides in individuals with eating disorders (for a recent in-depth review see Monteleone & Maj, 2013). Briefly, individuals with AN have state-dependent altered levels of CRH (Licinio, Wong, & Gold, 1996), neuropeptide-Y (NPY), beta-endorphin, and leptin that normalize with weight restoration (Bailer & Kaye, 2003; Kaye, 2008), whereas individuals with BN demonstrate state-related reductions in cholecystokinin (CCK) response (Brewerton, Lydiard, Laraia, Shook, & Ballenger, 1992; Hannon-Engel, 2012; Kaye et al., 1987; Lesem, Berrettini, Kaye, & Jimerson, 1991) and beta-endorphin levels. A number of the CNS neuropeptides implicated in AN and BN are also involved in regulating cognitive functioning, mood, the autonomic nervous system, and hormone secretion (Jimerson & Wolfe, 2006). While abnormalities in neuropeptide systems typically remit following recovery from AN and BN, malnutrition in combination with neuropeptide alterations can exaggerate symptoms of increased satiety and dysphoric mood, which might perpetuate eating disordered behavior (Bailer & Kaye, 2003; Monteleone & Maj, 2013); see Bailer and Kaye and Monteleone and Maj for full reviews of how neuropeptides influence AN and BN. Neuropeptides are also implicated in BED, and both human and animal studies suggest that binge eating alters the endogenous opioid system (Bencherif et al., 2005; Blasio, Steardo, Sabino, & Cottone, in press; Munsch, Biedert, Meyer, Herpertz, & Beglinger, 2009). Individuals with BED have higher meal-induced levels of CCK and PYY than controls (Munsch et al., 2009). Furthermore, both obese and nonobese women with binge eating demonstrate decreased levels of ghrelin in the morning, compared with nonobese healthy women and obese non-binge-eating women (Monteleone et al., 2005). However, these findings have not been consistently replicated across studies (Geliebter, Hashim, & Gluck, 2008; Munsch et al., 2009). The monoamine system, including serotonin (5-HT), dopamine (DA), and norepinephrine (NE), has also been implicated in the development and maintenance of disordered eating (Hildebrandt, Alfano, Tricamo, & Pfaff, 2010; Kaye et al., 2009; Steiger, 2004; Vaz-Leal, Rodriguez-Santos, Garcia-Herráiz, & Ramos-Feuntes, 2011). The 5-HT system is critical in regulating appetite, anxiety, and impulse control (Fairbanks, Melega, Jorgensen, Kaplan, & McGuire, 2001), and the effects of 5-HT manipulation on eating behaviors have been demonstrated in both animal and human models (e.g., Blundell, 1986; Mancilla-Diaz, Escartin-Perez, Lopez-Alonso, & Cruz-Morales, 2002). Studies of individuals with eating disorders document alterations in 5-HT metabolism, receptor sensitivity, and transporter activity (Bailer et al., 2011; Frank & Kaye, 2005; Kaye, 2008). As a general trend, decreased 5-HT is associated with increased feeding (Brewerton, 1995), leading to the expectancy that AN would coincide with increased 5-HT. At first glance, individuals with AN appear to contradict expectation with regard to levels of 5-HT. Individuals with AN have significant reductions in cerebral spinal fluid 5-hydroxyindoleacetic acid (CSF 5-HIAA) compared with controls (Kaye et al., 2009), suggesting reduced 5-HT activity. However, CSF 5-HIAA levels are elevated following long-term recovery from AN (Kaye, 2008), indicating that AN may correspond to a primary state of increased 5-HT and that diminished 5-HT activity may be a result of malnutrition, rather than a trait-related feature. PET and SPECT have been used to investigate the role of the 5-HT1A and 5-HT2A receptors in AN (Bailer & Kaye, 2011). Although studies have not been entirely consistent, most have shown that both ill and weight restored individuals with AN have reduced binding of 5-HT2A (Bailer et al., 2004; Frank et al., 2002; Kaye et al., 2001) and increased binding of 5-HT1A (Bailer et al., 2005; Bailer et al., 2011). In an animal model, interactions between 5-HT1A and 5-HT2A in the medial prefrontal cortex have been implicated in anxiety, attention, impulsivity, and compulsive behavior (Carli, Baviera, Invernizzi, & Balducci, 2006; Krebs-Thompson & Geyer, 1998; Winstanley et al., 2003). This is an interesting finding given that these traits have been implicated in AN, and particularly AN binge-purge type (AN-BP). Findings in acute BN are generally compatible with a low 5-HT hypothesis (decreased 5-HT promotes increased feeding). Individuals with BN demonstrate decreased CSF 5-HIAA levels (Kaye, 2008) that are inversely related to binging and purging frequency (Jimerson, Lesem, Kaye, & Brewerton, 1992), reduced platelet binding of 5-HT uptake inhibitors, reduced availability of central transporters, and decreased neuroendocrine responses to 5-HT precursors and 5-HT agonists/partial agonists. However, similar to individuals with AN, following recovery, they have elevated levels of CSF 5-HIAA compared to controls (Kaye, 2008). Abnormalities in 5-HT have also been implicated in binge eating (Akkermann, Nordquist, Oreland, & Harro, 2010) and in the frequency of binge eating for individuals with BN (Jimerson et al., 1992; Monteleone, Brambilla, Bortolotti, Ferraro, & Maj, 1998). Decreased 5-HT responses are hypothesized to contribute to blunted satiety, which may increase propensity for binge eating (Chiodo & Latimer, 1986). Further support of the role of 5-HT in eating disorders is provided by studies indicating that selective serotonin reuptake inhibitors (SSRIs) are fairly efficacious in treating BN and BED (see Brownley, Berkman, Sedway, Lohr, & Bulik, 2007; Shapiro et al., 2007 for reviews). Fluoxetine is the only FDA-approved medication for the treatment of any eating disorder. Fewer investigations have examined the effectiveness of SSRIs in treating AN, and in the small number of available studies (Attia, Haiman, Walsh, & Flater, 1998; Ferguson, La Via, Crossan, & Kaye, 1999; Kaye et al., 2001; Rosenblum & Forman, 2003; Vaswani, Linda, & Ramesh, 2003; Walsh et al., 2006), results were mixed. In summary, significant evidence suggests an overall dysregulation of 5-HT in eating disorders (Steiger, 2004; Kaye et al., 2013), which persists following recovery. Together, these results suggest that patterns of 5-HT dysregulation might vary by eating disorder subtype, suggesting that underlying pathophysiology might differ across varying eating disorders presentations (Bailer et al., 2013; Kaye, 2008). Dopamine is another monoamine hypothesized to contribute to disordered eating (Bailer et al., 2013; Bello & Hajnal, 2010; Frank & Kaye, 2005; Jimerson et al., 1992; Kaye, Frank, & McConaha, 1999), and it is known to be involved in the reward and motivational aspects of feeding behavior (Erlanson-Albertsson, 2005; Szczypka, Rainey, & Palmiter, 2000). Individuals in recovery from restricting-type AN (AN-R) show lower CSF levels of the DA metabolite homovanillic acid (HVA; Kaye et al., 1999), which is typically considered an indicator of reduced dopamine function and reduced dopamine turnover (dopamine to HVA ratio). Individuals recovered from AN-R and AN-BP also demonstrate increased binding of D2/D3 receptors in the anteroventral striatum (AVS; Bailer et al., 2013; Frank et al., 2005), including the nucleus accumbens, a brain region implicated in the response to reward stimuli (Delgado, Nystrom, Fissell, Noll, & Fiez, 2000; Montague, Hyman, & Cohen, 2004). Individuals with BN, particularly those with high binge frequency, also have significantly lower HVA levels (Jimerson et al., 1992; Kaplan, Garfinkel, Warsh, & Brown, 1989; Kaye et al., 1990). However, Jimerson et al. (1992) found that after weight restoration and normalization of food intake, individuals recovered from BN did not differ significantly from controls on HVA concentrations, suggesting that abnormalities in the dopamine system in BN might be state dependent. Lastly, DA has also been hypothesized to play a role in binge eating disorder by modulating reward pathways (Bello & Hajnal, 2010; Mathes, Brownley, Mo, & Bulik, 2009). Norepinephrine (NE) transmission in the medial prefrontal cortex is also implicated in food-related motivational behavior in animal models (Ventura, Latagliata, Morrone, La Mela, & Puglisi-Allegra, 2008; Ventura, Morrone, & Puglisi-Allegra, 2007). Although few investigations have specifically examined the role of NE in disordered eating, it plays a central role in CNS modulation of energy balance, which has downstream effects on satiety, hunger, and feeding behavior (Hainer, Kabrnova, Aldhoon, Kunesova, & Wagenknecht, 2006). Overall, the field is embracing more complex systems and pathway-driven models of disease to understand the complicated way in which monoamines and other neurotransmitters are implicated in disease etiology (Kaye, 2008; Kaye et al., 2009). Further, there is evidence to suggest that these neurotransmitter systems likely act in concert, contributing to behaviors associated with disordered eating. For example, a recent study by Bailer et al. (2013) using PET showed that an interaction between the 5-HT transporter and striatal DA D2/D3 receptor radioligand binding measures was associated with harm avoidant symptoms in women recovered from eating disorders. Based on this finding, authors hypothesize that interactions between the 5-HT and DA systems may contribute to eating disorder symptoms. Neuroimaging studies with CT show neuroanatomical changes in individuals with AN, including cerebral atrophy and enlarged ventricles (Artmann et al., 1985; Heinz et al., 1977; Krieg et al., 1989; Lankenau, Swigar, Bhimani, Luchins, & Quinlan, 1985; Nussbaum, Shenker, Marc, & Klein, 1980; Titova, Hjorth, Schiöth, & Brooks, 2013). A 2012 systematic review by Van den Eynde et al. reported that the eight studies they included found reduced gray matter volume in AN in the insula, frontal operculum, and occipital, medial temporal, or cingulate cortex. MRI studies in AN also demonstrate increased volumes of CSF in association with deficits in both total gray matter and total white matter volumes (Castro-Fornieles et al., 2010; Joos et al., 2010; Katzman et al., 1996; Titova et al., 2013) and enlarged ventricles (Golden et al., 1996). There is much debate over whether these changes persist after successful treatment and weight restoration. Several investigations have reported that neuroanatomical changes persist following normalization of weight (Artmann et al., 1985; Krieg et al., 1989), whereas other studies have found that brain tissue may increase with weight restoration in AN (Roberto et al., 2011) and that structural brain abnormalities are reversible after long-term recovery (Golden et al., 1996; Wagner et al., 2005) but not short-term recovery (Friedrich et al., 2012). Although definitive conclusions cannot be drawn, if lasting brain abnormalities in AN do occur, they might represent residual damage to the brain or persistent abnormal metabolism (Husain et al., 1992). They could also represent underdeveloped areas that originally contributed to the eating pathology (Artmann et al., 1985). Findings regarding structural abnormalities are mixed (Frank, 2013). Several studies support structural changes in BN, including cerebral atrophy and decreased ventricle size (Hoffman et al., 1989; Krieg et al., 1989). Other studies suggest normal or increased localized gray matter in the orbitofrontal cortex and striatum (Joos et al., 2010). A 2010 study by Schäfer, Vaitl, & Schienle, found that individuals with BN had greater medial orbitofrontal cortex volume relative to controls. Further, in individuals with BN who had increased ventral striatum volumes, purging severity and BMI were correlated with striatal gray matter volume, suggesting a potential correlation between behavioral and neuroanatomical findings. Reductions in inferior frontal regions correlated inversely with symptom severity, age, and Stroop interference scores in the BN group. Marsh et al. (in press) reported significant reductions, in 34 adolescent and adult patients with BN relative to healthy controls, of local volumes on the brain surface in frontal and temporoparietal areas in the BN participants. The authors suggested that this difference could be related to deficits in self-regulation seen in BN. Other studies, however, have found no evidence for neuroanatomical abnormalities in BN (Husain et al., 1992; Joos et al., 2010). Taken together, results from these studies indicate that neuroanatomical abnormalities often occur in individuals with eating disorder, particularly AN. Additional prospective studies are needed to understand whether these abnormalities are a cause or an effect of the disordered eating behavior. As noted previously, biology only accounts for part of the liability to developing an eating disorder. It is hypothesized that environment, via channels such as learning, modeling, and life events, can contribute to eating disorders risk either directly or indirectly through their influence on genetic expression. Stressful life events have long been hypothesized to play an important role in eating disorder etiology (Klump, Wonderlich, Lehoux, Lilenfeld, & Bulik, 2002; Pike et al., 2006; Schmidt, Troop, & Treasure, 1999). However, research in this area is fraught with methodological challenges. Many studies have included exclusively clinical samples of individuals with eating disorder symptomatology, did not include controls, and assessed life events retrospectively (Berge, Loth, Hanson, Croll-Lampert, & Neumark-Sztainer, 2011; Raffi, Rondini, Grandi, & Fava, 2000; Schmidt et al., 1999). Nonetheless, one investigation that did include a community-recruited sample of women with BN and matched controls suggested that individuals with BN were more likely than controls to experience certain stressful life events (e.g., a major move, illness, pregnancy, physical abuse, and sexual abuse) during the year prior to the beginning of their illness (Welch, Doll, & Fairburn, 1997). There was no association between BN status and the occurrence of other life events (e.g., bereavement, illness of a close relative, friend, or partner, and beginning or ending a romantic relationship) in the last year. In addition, 29% of women with BN experienced none of the life events assessed in the 12 months prior to the onset of their diagnosis.
Feeding and Eating Disorders
Description of the Disorders
Clinical Picture
Diagnostic Considerations
Epidemiology
Psychological and Biological Assessment
Structured Interviews
Self-Reports
Medical Assessment
Etiological Considerations
Behavioral Genetics and Molecular Genetics
Neuroanatomy and Neurobiology
Neuropeptides
Monoamines
Structural Abnormalities
Learning, Modeling, and Life Events
Life Events
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
