Bulimia nervosa (BN) is an eating disorder with 5 key characteristics as noted by the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DSM-5-TR).[1]
Severity can be mild with 1–3 episodes, moderate with 4–7 episodes, severe with 8–13 episodes, or extreme with 14 or more episodes per week.
Patients with BN may experience symptoms in the following areas:[2, 3, 4] Specific symtpoms are discussed in Presentation.
The diagnosis is usually made by assessment of reported behavior, but laboratory studies are helpful to support the diagnosis and, importantly, assess severity of the condition and determine level of care.
Lab studies that may be used for diagnosis include:[7]
Electrocardiography
Because of the potential for arrhythmias and cardiomyopathy as possible complications of BN, an electrocardiogram (ECG) should be performed in patients who are very thin, complaining of palpitations, or have other signs or symptoms of cardiovascular concern.[6]
DEXA
Because of the potential for osteoporosis, a dual-energy radiographic absorptiometry (DEXA) scan may be useful, particularly for patients with irregular menses, who have a history of anorexia nervosa, who have mood disorders, and/or who smoke cigarettes.
See Workup for more detail.
Imaging
Imaging studies are not routinely indicated or ordered for uncomplicated or typical cases of BN. Nonetheless, there has been growing interest in the use of neuroimaging techniques to explore the structural and functional brain changes that take place in those with eating disorders, mostly focusing on patients with anorexia nervosa, but now starting in BN.
See Workup for more detail.
Patients with BN are typically treated in an outpatient setting. However, at times there is indication for medical or psychiatric inpatient stabilization. See Treatment for more detail.
Bulimia nervosa (BN) is an eating disorder delineated in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DSM-5-TR).[1]
BN includes regularly occurring compensatory behaviors that are intended to rid the body of the excess calories consumed during eating binges. BN is distinguished from the recently established binge eating disorder (BED), in which no regular or consistent compensatory behavior accompanies the binging episodes. Although the DSM-5-TR describes that compensatory behaviors may include purging and/or non-purging behaviors (see below), in contrast to earlier DSM editions, DSM-5-TR no longer maintains specific purging and non-purging subtypes.
In contrast to individuals with uncomplicated BED who tend to be obese, people with BN are usually within the normal weight range. Binge eating and compensatory behaviors can also occur in anorexia nervosa (AN), but the distinguishing characteristic is that individuals with AN are underweight. The natural history of eating disorders is such that individuals may pass through several diagnoses over time with some degree of diagnostic crossover, with about 27% of individuals with BN developing AN and 35% with AN developing BN at various time points.[141]
BN is characterized by frequent episodes of binge eating associated with emotional distress and a sense of loss of control.
Binge eating is eating, in a discrete period of time (eg, 2 hours) an amount of food that is significantly larger than is typical for most people during the same defined period. This behavior is associated with a perceived loss of control of eating during this time.
Binge eating episodes have to be distinguished from overeating episodes (consumption of an unusually large amount of food in a defined period, without concomitant perception of loss of control) or subjective bulimic episodes (consumption of objectively normal or minimal amounts of food in a defined period with a perception of loss of control).
Compensatory behaviors used by individuals with BN include self-induced vomiting, laxative abuse, excessive exercise generally experienced as being joyless and/or compulsive, episodes of fasting or strict dieting, diuretic abuse, use of appetite suppressants, failure to use insulin in those with type I diabetes, and/or the use of medications intended to speed up metabolism (eg, thyroid hormone, stimulants).
DSM-5-TR diagnostic criteria require that the binge eating and inappropriate compensatory behaviors both occur, on average, at least once a week for three months.
Individuals with BN are dissatisfied with their body shape, weight, or both and usually show some degree of body image distortion (believing one looks much fatter than is actually the case). These are accompanied by associated abnormalities in mood and in perceptions of hunger and satiety. Self-esteem is frequently most prominently determined by self-perceived shape and weight.
A 21-year-old woman is brought into an outpatient clinic by her mother, who complains that her daughter has been demonstrating unusual eating patterns since she moved back home 6 months ago. Her mother observes her to eat large amounts of food, such as desserts, when she is alone, often finding food wrappers hidden in her daughter’s room. She is worried that her daughter may be engaging in vomiting after these episodes of heavy eating. She often isolates herself in the bathroom for 10–20 minutes after a large meal. This case will be discussed in the Clinical History section.
Binge eating disorder (BED)
BED is characterized by frequent and recurrent binge eating episodes without consistent compensatory behaviors. It is the most common eating disorder overall, with a lifetime prevalence of 2.8%.[10] BED is known to have several comorbidities with other psychiatric disorders, and is strongly associated with obesity.[142] Lifetime prevalence estimates of BED are 3.5% among women and 2.0% among men.
Other specified feeding or eating disorders
Those syndromes include:
The pathophysiology of bulimia nervosa (BN) remains unclear. Abnormal laboratory values typically normalize with regulation of eating behaviors and cessation of binge eating and purging behaviors.
There are no animal models for BN but more broadly for binge eating behaviors, indicating acute elevations in dopamine neurotransmission but decreases in striatal dopamine receptor distribution after prolonged binge eating and anxiety when deprived of binge eating.[152, 153, 154] Animal models have suggested a change in the balance of dopamine D1 and D2 receptor neurotransmission with disordered eating, affecting midbrain serotonin neurotransmission and indicating state-dependent inter-relationships between neurotransmitter systems.[155] Sex hormones and neuroactive peptides are also frequently altered during the ill state of eating disorders and affect brain neurotransmission,[156] as do the fat-cell-derived hormones leptin or ghrelin from the gastric mucosa that may stimulate or dampen brain dopamine response and alter food approach in eating disorders.[157, 158] In summary, animal studies suggest distinct changes, especially in serotonin and dopamine neurotransmission, associated with food restriction or binge eating in eating disorders.
Human neurotransmitter receptor studies using positron emission tomography (PET) showed higher serotonin 1A-receptor binding in AN and BN when ill and after recovery, suggesting state-independent alterations. Ill BN also showed no dopamine D2-receptor binding differences versus controls, but lower striatal dopamine release was associated with higher binge-eating frequency.[159] Recently, higher glutamate receptor binding across several regions that was related to maturity fears characterized BN compared to controls.[160] However, studies on GABA or glutamate using MRI spectroscopy in the ED population have yielded inconclusive results.[161]
Altogether, while the literature is limited, the studies indicate that, in particular, serotonin receptor availability is altered in AN and BN, making those receptors potential drug intervention targets. However, those neuroreceptor studies did not inform on the functionality of those receptors.
Task-based functional magnetic resonance imaging (fMRI) studies measure brain activation during tasks that test specific behaviors and allow to associate behavior with regional brain response. Those studies typically do not directly involve neurotransmitters, but regional activation and task specificity can allow inference on neurotransmitter circuits triggered.
Salience network
One of the brain circuits that have been most reliably associated with neurotransmitter function is the reward or salience network. Food is a natural salient stimulus, and reward pathways similar to substances of abuse are activated when we desire, approach, or eat food, but also when we try to avoid salient stimuli.[162, 163] Important regions in this circuitry include the ventral striatum (receives midbrain dopaminergic input, processes motivational salience), orbitofrontal cortex (stimulus valuation), and anterior cingulate (error monitoring, reward expectation).[164] The primary neurotransmitters involved are dopamine for stimulus approach and avoidance and opioids for processing hedonic experiences.[164, 165] A paradigm closely associated with brain dopamine response is the prediction error model, a Pavlovian conditioning paradigm where individuals learn to associate unconditioned taste or monetary with conditioned visual stimuli.[166] This paradigm tests the brain reward valence (processes whether expectations are violated) and motivational salience networks (processes stimuli that propel an individual's behavior towards or away from a particular object, perceived event, or outcome).[167] A large transdiagnostic study indicated a strong inverse relationship between the dopamine-anchored prediction error response and body mass index (BMI) across AN, BN, and BED groups. Furthermore, that study indicated a positive relationship between the prediction error signal and activation of the food control circuitry, suggesting that responsiveness of dopamine circuits is associated in AN with higher, but in BN and BED, lower food control circuit function.[168] This hypothesis was supported by a study that paired a monetary reward (incentive delay) task during fMRI with a catecholamine depletion paradigm suggesting desensitized dopamine function in BN.[169] Increased ED symptoms during catecholamine depletion in BN after recovery further suggested dopamine or noradrenaline as vulnerability factors.
In BN, negative affect correlated positively with striatal brain response during milkshake receipt.[170] Low mood may enhance the reward value of food stimuli in BN and trigger binge eating. Others showed lower frontal cortical, ventral striatal, and hippocampal activation in BN that correlated with binge/purge frequency in a task that provided monetary reward when navigating through a maze.[171, 172] It was hypothesized that altered learning, executive control, and reward response could interact and be effects of both abnormal brain development and consequences of illness behavior.
In summary, the reward or salience network, including pharmacological interventions, is a prime target for ED research. The transdiagnostic association of dopamine-related brain response with BMI and with food control circuits makes this circuitry an excellent target to manipulate or modulate using medication.
Cognition, anxiety, and stress
Individuals with BN showed worse cognitive performance when stressed, but showed improved cognitive performance in response to positive emotions and associated with altered striatal activation compared to controls.[173, 174] Individuals with EDs tend to be anxious, with anxious traits as potential risk factors for EDs, suggesting interactions between serotonin (mood, anxiety) and dopamine (cognitive flexibility, salient stimulus-response) pathways.[175, 176, 177, 178] We hypothesize that extremes of food restriction or overeating alter dopamine-related brain response, and anxious conditioning to food intake, recruits those circuits to engage in fearful avoidance as opposed to food approach.[179, 180]
A literature review of the prevalence of eating disorders worldwide in young people suggests that bulimia nervosa (BN) has a lifetime prevalence of between 0.8% and 2.6% in women and between 0.1% and 0.2% in men.[143] Most estimates of the prevalence of BN and other eating disorders come from Australia, Western Europe, and North America. However, smaller-scale epidemiological estimates suggest a similar prevalence of BN in many other regions and countries around the world, including Malaysia and Russia.[144, 145] Older estimates found lifetime prevalence estimates of BN (DSM-IV) to be 1.5% among women and 0.5% among men. Retrospective age-of-onset reports suggested that risk of BN increased with successive birth cohorts and that BN was significantly comorbid with many psychiatric disorders.[142]
BN was found to be more common among those whose occupation or hobbies require gaining and/or losing weight rapidly, such as wrestlers and competitive bodybuilders.[13] Athletes in certain sports (eg, runners and gymnasts) are particularly prone to eating disorders.[14] Athletes' most frequent hindering factors were negative emotions/cognitions, sport pressures, and hurtful modeling, while non-athletes reported negative emotions/cognitions, lack of support, and hurtful modeling. The female athlete triad of eating disorders, hypothalamic amenorrhea, and osteoporosis is now well recognized and is particularly common in sports where slimness and body shape are of great importance, such as gymnastics, long-distance running, diving, and figure skating. However, the presence of the female athlete triad should suggest a diagnosis of anorexia, either typical or atypical. Eating disorders are also being recognized as a problem in predominantly male sports such as cycling, weight lifting, and wrestling. Certain vocations such as acting, modeling, and ballet dancing[15] also appear to be associated with higher risk for these disorders.
BN is a cosmopolitan disorder that has been described in all ethnic, racial, and socioeconomic groups. The literature is mixed regarding ethnic differences in eating disorders. No clear consensus exists about the relative prevalence of eating disorders and associated symptoms across ethnicities. Clinicians should thoroughly evaluate all patients, regardless of race or ethnic background, who are presenting with signs and symptoms suggestive of an eating disorder.[18, 19]
In the US National Comorbidity Survey Replication, lifetime prevalence was consistently 1¾ to 3 times as high among women as men for the BN as in other eating disorders.[142] Clinicians should remain aware that men also develop BN and other eating disorders, but that males with BN are often not diagnosed or treated for long periods of time.[149] The psychopathology and attitudes of males with eating disorders appear on the whole to be similar to those of females with eating disorders; both are significantly associated with family histories of these disorders. Although few data are available, evidence suggests that men and women also share significant similarities in clinical course, complications, and response to treatment.[150]
The median age of onset of BN is 18 years, similar to the median age of onset for AN.[151] BN has also been reported in the elderly.[20]
A meta-analysis of 36 studies by Arcelus et al suggested that individuals with eating disorders have significantly elevated mortality rates, with standard mortality rates (SMR) about 2 times higher in bulimia nervosa (BN) patients than in controls.[17] However, another study looking at registry of all deaths among eating disorder patients in the University of Minnesota healthcare system found that mortality rates were around 4% for BN and comparable to those for anorexia nervosa (AN).[146] Patients with eating disorders overall have about 50% higher healthcare costs per year compared to age-matched controls.[147] A cohort study from Quebec, Canada suggests that increased risk of sudden cardiac death and cardiovascular events may account for a large proportion of the increased mortality in persons with BN. This cohort study found that BN was found to be associated with ischemic heart disease (HR, 6.63; 95% CI, 3.34–13.13), atherosclerosis (HR, 6.94; 95% CI, 3.08–15.66), and cardiac conduction defects (HR, 2.99; 95% CI, 1.57–5.71). BN was also associated with 21.93 (95%CI, 9.29-51.74) times the risk of myocardial infarction at 2 years of follow-up and 14.13 (95% CI, 6.02–33.18) times the risk at 5 years of follow-up.[148]
Research to date suggests a variable prognosis for bulimia nervosa (BN). The illness may pursue a long-term, fluctuating course over many years, or may be more episodic, associated with stressful life events and crises. The diagnosis may not be stable over time.[22] In the shorter term, some reports suggest a 50% improvement in binge eating and purging behavior among patients who are able to engage in treatment. In a 12-year outcome study that looked at BN, purging type, 28.2% of the individuals maintained the diagnosis of BN. Psychiatric comorbidities predicted poor outcome, specifically self-injurious behaviors.[23]
In 2008, a 10-year follow-up study was published that looked at parental psychopathology as a source of predicted outcome. The paper found that substance abuse in fathers and depression in mothers was associated with poor outcome. Obesity in mothers was associated with a better long-term outcome.[24]
In another study that examined temporal patterns of recovery in BN, 10% of those with BN met recovery criteria at 10 years. At 15 years, 25% met recovery criteria. The patients had 3 times the rate of recovery at 10–14 years than matched patients with anorexia nervosa (AN).[10] In a 5-year longitudinal study, patients with BN had a remission rate of approximately 74% and a relapse rate of approximately 47%. The natural course did not appear to be influenced by personality disorder psychopathology.[26]
Consistent predictors of outcome have not yet been identified. However, the severity of the purging sequelae, negative self-image,[27] childhood maltreatment,[28] childhood obesity/overeating,[29] individual/family eating patterns during childhood/early adolescence,[30] and ADHD[31] may be important indicators of worse prognosis. Interestingly, a longitudinal cohort of patients with AN-R, AN-BP, and BN over 22 years did not find that depression was a statistically significant predictor of either recovery or not recovering from bulimia.[182] Electrolyte imbalances, esophagitis, and hyperamylasemia reflect more severe purging and may predict a poorer outcome, due to higher risk of dying.
A cohort study that assessed women with AN and BN 9 years and then 22 years after the initial diagnosis found an unchanged majority of the women with BN (68.2% at both timepoints) did recover. This cohort study found that early recovery was not predictive of long-term prognosis in women with BN.[182] In this same 9-year and 22-year cohort 9-year longitudinal study, when compared to women with BN who have no history of AN, patients diagnosed with BN and with a history of AN were more likely to cross back into AN and were less likely to achieve full recovery at 9 years.[32] Therefore, lifetime history of AN may be an important indicator of worse prognosis, at least earlier on in the disease course, in patients with BN. Importantly, the 22-year followup of this cohort showed almost identical rates of recovery for AN and BN, with about two-thirds of both groups having recovered at that timepoint.[182]
Cognitive behavioral therapy (CBT) remains the therapeutic method of choice for bulimia nervosa (BN), and various modifications of this technique are actively under investigation. Most of these interventions include the premise that education about BN in a nonthreatening environment has a therapeutic effect. These types of therapy are conducted in either individual or group settings. Educational components of treatment address the following issues:
More than 70% of published management studies of BN involve some form of psychoeducational program. Although no “unbundling” studies have been conducted that exclude psychoeducation to assess the relative contribution of this specific strategy to overall treatment outcomes, anecdotal reports and the personal experiences of many practitioners suggest that for at least some patients the educational information helps significantly.
Family members can provide perspectives on factors contributing to the onset of the disorder and issues that may help or hamper recovery efforts, and their involvement is often critical to sustained recovery. In addition to empathically listening to family members, clinicians should educate and advise them on the nature of the disorder and their interactions with the patient. When indicated, and with the patient’s consent, families should be involved in treatment. Such involvement may contribute to the likelihood of better outcomes.[8]
For further information, see the following websites:
Following is a list of workbooks and books for BN:
Other books reported to be helpful by patients/families include the following:
Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DMS-5-TR) diagnostic criteria for bulimia nervosa (BN) are as follows:[1]
Specify if either of the following applies:
Specify current severity. The minimum level of severity is based on the frequency of inappropriate compensatory behaviors. The level of severity may be increased to reflect other symptoms and the degree of functional disability. Specify current severity as follows:
Bulimia nervosa (BN) is often not diagnosed for many months or even years after onset because of patients' secretiveness about their symptoms, usually associated with a great deal of shame. These patients often see physicians for other problems, such as anxiety, depression, infertility, bowel irregularities, fatigue, or palpitations. Similarly, they may see mental health professionals for mood and anxiety problems, personality issues, relationship issues, histories of childhood or adolescent trauma, or substance abuse, without revealing the presence of an eating disorder.
One common presenting scenario is that of a patient who is concerned about their weight who seeks help with weight loss. Symptoms may include bloating, constipation, and menstrual irregularities. Far less often, people may present with palpitations resulting from arrhythmias, which are often associated with electrolyte abnormalities and dehydration. BN is also often, but not always, characterized by an inappropriate premium placed on slender physical appearance. It is important to note that the diagnostic criteria do not necessarily require an individual to express a desire to lose weight or change their appearance to meet criteria for BN. Some, especially younger patients, may not be able or willing to articulate why they engage in eating disorder behaviors. Others may express a motivation for control, a fear of fullness, or other reasons for their disordered eating behaviors. Failure to make an appropriate eating disorder diagnosis that an individual would otherwise meet criteria for, whether it be anorexia nervosa (AN), BN, or other specified feeding or eating disorder (OSFED), because of a lack of expressed concern or denial of concern about body weight/shape is dangerous for patients’ physical and mental health due to causing delays in diagnosis that can contribute to significant morbidity and mortality.
A dietary history may reveal attempts to control weight by dieting and abstaining entirely from high-calorie foods at all times except during binge eating episodes. Often, an all-encompassing preoccupation with food and eating is present, and recurring cycles of extreme dieting and/or fasting may alternate with gorging behavior. The clinician should inquire further about types of foods and quantities of foods consumed when a patient endorses binging behavior. A binge eating episode, by definition, must constitute consumption of more food than would typically be consumed in a single setting by an individual, so there is some subjectivity in what is a binge. Some patients with AN or with periods of very restrictive eating patterns may endorse a “binge,” but on further inquiry may be describing an episode that would be either less than or normative for typical eating for a non-eating-disordered individual but perceived subjectively as a binge eating episode. In the context of highly restrictive eating, any instance of normative eating, such as consuming a dessert, eating high-calorie snacks, or even allowing oneself to eat until feeling full, might appear to be binge eating to an individual who is used to eating very little amounts of food most of the time.
Most patients self-induce vomiting by gagging themselves with their fingers or a toothbrush. Some patients are able to regurgitate reflexively, without requiring external stimulation of the pharynx.[34] A minority of patients will chew, then regurgitate, without actually swallowing the food. One particularly dangerous form of vomiting is via induction through the use of emetics (eg, ipecac). Ipecac is a tightly binding and slowly released mycotoxin that may lead to fatal cardiomyopathy in habitual users. Up to 40% of patients misuse laxatives, thinking that their use will help them lose weight. In fact, laxative misuse results in additional dehydration and often electrolyte abnormalities as well. Screening for laxative use should be a routine topic of assessment.
For children and adolescents, the parents/guardians of the patient should be asked to check for signs of vomiting and laxative misuse in the home when the patient is denying purging behavior but there is a strong suspicion of purging behaviors occurring. Indications of surreptitious purging behaviors that parents/guardians should be instructed to look for in the home include hidden vomitus in concealed places in the home (eg, closets, floorboards), concealed laxatives in the patients’ room or belongings, clogged drains with vomit, vomit in trash cans, and so on. The parents/guardians should also be instructed to be observant for other signs of purging behavior both for diagnostic and preventative purposes such as unusual online orders (which could be for laxatives or other emetics) and unusually long periods of time spent in the bathroom or shower, which may or may not be after meals and may or may not be accompanied by retching noises. For both children and adults, collateral informants can be very useful sources of information for the clinician about surreptitious purging or other eating disorder behaviors.
Clinicians should be aware of misconceptions that eating disorders are almost exclusively present in young women to avoid missing diagnoses of BN or other eating disorders in male patients or older patients. Additionally, it is a misconception that eating disorders, including BN, mostly occur in White women and people from Western cultures. While research is still evolving on the epidemiology of BN and other eating disorders in males, older individuals, people in developing countries, and so on, in the authors’ clinical experiences, we have encountered a great many patients with BN and other eating disorders who are male, not White, and of a great many different ethnic and racial backgrounds.
Patients with BN may experience the following symptoms:
A high index of suspicion is required in any depressed or anxious weight-conscious young woman.
A set of screening questions, such as the SCOFF mnemonic questionnaire,[35] is useful to obtain a quick impression as to the potential need for further in-depth questioning. The SCOFF questionnaire includes the following 5 questions:
The Eating Disorder Screen for Primary Care (ESP) questionnaire is an alternative screening tool.[36] It contains the following 5 questions:
The Eating Attitudes Test (EAT) is a self-report population-based screening instrument that patients can complete in the waiting room prior to seeing the health care provider.[37] (See Psych Central for more information.)
Family history of eating disorders, anxiety, mood disorders, and alcohol and/or substance abuse/dependence may contribute to the risk of BN and should be investigated.
Many physiological abnormalities may be seen in association with eating disorders, but virtually all appear to be consequences of the abnormal behaviors, not their causes. In most cases of BN, laboratory abnormalities are relatively minor. In cases of very frequent purging (eg, daily or multiple times per day), abnormalities in electrolyte and serum amylase levels occur, but these and most other laboratory abnormalities are reversible with weight restoration and cessation of compensatory behaviors.
Among the identified metabolic consequences sometimes seen in BN are low plasma insulin, C peptide, triiodothyronine, and glucose values, as well as increased beta-hydroxybutyrate and free fatty acid levels. Both fasting and post-binge/post-vomiting hypoglycemia are sometimes seen in some patients with BN. Some studies suggest increased secretory diurnal amplitudes in cortisol and adrenocorticotropic hormone (ACTH) in BN as well as blunted responses to corticotrophin-releasing hormone (CRH). However, these findings have been inconsistent among research studies.
Reports have also suggested abnormal responses to dexamethasone suppression like those seen in AN and major depressive disorder, more common among individuals with significant dietary restriction. Some authors have attributed these abnormalities to impaired dexamethasone absorption, which is demonstrated in some patients with BN. Similar to findings in AN, patients with BN tend to have higher growth hormone levels at night, while nocturnal prolactin levels tend to be less than those seen in controls.
About half of women with BN have anovulatory cycles, while about 20% have luteal phase defects. Patients with anovulatory cycles generally have impaired luteinizing hormone pulsatile secretion patterns and associated reduced estradiol and progesterone pulse amplitudes.[39, 40]
Although the implications of many research findings are still unclear, and none of the following offer clinical tests of any merit, reports suggest involvement of the serotonin transporter,[41, 42] autoantibodies against neuropeptides,[43] various chromosome regions,[44] brain-derived neurotrophic factor,[45] and peptides leptin and ghrelin.[46] In a few instances, cerebral hemispheric lesions may be involved in pathogenesis.[47] Regional cerebral blood flow abnormalities have been noted in adolescents.[48] Endogenous opioids and beta-endorphins have been implicated in the maintenance of binge eating. Therefore, diagnosis of BN and other eating disorders should be made mostly on the basis of interview with the patient and their collateral informants as well as behavioral observation at times (eg, purging witnessed by family or a staff member). Laboratory tests and physical examination can provide important information about comorbid medical sequelae of BN, but diagnosis requires a thorough diagnostic interview and clinical history.
The underlying causes for buimia nervosa (BN) remain elusive. However, a variety of biological and psychological factors have been suggested to be involved the development of BN.
BN has long been associated with inadequate mechanisms to control food intake beyond ones physiological needs, and behavioral traits could contribute.[183, 184, 185, 186, 187]
Emotion regulation has been defined as the “extrinsic and intrinsic process responsible for monitoring, evaluating, and modifying emotional reactions, to accomplish one’s goals."[188] A disturbance in emotion regulation has been found in eating disorders.[189] Individuals with BN have difficulties modulating strong emotions and controlling rash, impulsive response.[190] Most but not all studies suggest that negative affect precedes binge eating episodes,[191] followed by initial relief.[192, 193]
Impulsivity, the “opposite to aspects of executive function,”[194] is a tendency to act with insufficient forethought, or a predisposition toward rapid, unplanned reactions to internal or external stimuli without regard for the negative consequences of these reactions. Impulsivity has relevance for binge eating behaviors, as those episodes typically occur impulsively in response to an external or internal trigger. Increased impulsivity had been found across BN.[195, 196] Overall, the literature on executive function in BN is limited, but there is evidence that altered impulsivity and executive function distinguish binge eating individuals from normal weight or obese controls.
Negative urgency, the tendency to experience strong impulses under the influence of negative emotions, or to act rashly when distressed, is related to emotion regulation and impulsivity and has been associated with binge eating.[197, 198] Negative urgency is a trait that is triggered by negative affect and may result in binge eating in vulnerable individuals.[199, 200, 201, 202, 203, 204] Interestingly, negative urgency and affect seem to be more relevant than impulsivity, at least for some with binge eating.[205]
Sensitivity to reward and punishment described in the reinforcement sensitivity theory (RST) provides a framework for how differences in brain systems' responsiveness to reward and punishment are reflected in individual personality.[206, 207] Reward sensitivity influences decision making in eating disorders.[208] Sensitivity to reward was found elevated in BN and it has been hypothesized that an imbalance between reward sensitivity, impulsivity and inhibition are mechanistically involved in binge eating.[209, 210, 211] In addition, individuals with binge eating behaviors showed greater risk taking, and obese binge eating disorder has been associated with altered value computation and discrimination of salient stimuli.[212]
Taken together, emotion regulation, impulsivity, negative affect, negative urgency, and sensitivity to reward have been linked to binge eating and may create a vulnerability for developing or perpetuation BN behaviors.[213] However, we do not have a transdiagnostic model for their underlying neurobiology.
A complex interplay exists between emotion regulation and cognitive control. Emotions affect attention, drive cognitive bias, and may interrupt proper decision making; on the other hand, attention to specific goals can control emotions and override strong feelings.[214, 215, 216] Control of food cravings is thought to involve prefrontal cortical areas, whereas greater caloric intake has been related to higher activation in gustatory cortex and brain regions for reward computation.[217, 218, 219] Little is known about how emotion regulation and cognitive control circuits affect binge eating. One study found reduced prefrontal cortical activity in BN when viewing food pictures.[220] BN showed hypoactivity in brain areas involved in self-regulation and impulse control, such as the prefrontal cortex or insula.[221, 222, 223, 224] Only one study directly investigated negative affect in relation to binge eating in BN, finding a positive correlation between negative affect and striatal brain response during anticipation of a milkshake.[225] Altogether, studies suggest altered brain function related to emotion regulation in individuals with binge eating, but the literature is inconsistent.
Animal models have shown that dopamine, serotonin, acetylcholine, and norepinephrine have all been associated with cognitive control and impulse control in frontal cortical circuits.[226, 227] For instance, dopamine D2 and serotonin 2A and 2C receptor signaling can modulate impulsivity.[228, 229] Human studies in BN found that serotonin 1A or dopamine D2/3 receptor binding correlated with harm avoidance or behavior inhibition.[230, 231] Endocrine factors such as ghrelin, leptin, sex hormones, and cortisol can influence food intake behaviors,[232] but how they contribute to eating disorders is still elusive.[233, 234] Basic research on the gut–brain axis showed how leptin or ghrelin activate brain stem and ventral striatum to activate dopamine circuits and motivation to eat.[231, 234] Stress, via cortisol and gut hormone activation, leads to dopamine-mediated decreased food intake in animals, although in humans stress often leads to increased food intake, suggesting a different pathophysiology.[235] Dopamine drives motivation and food approach, while hedonic aspects of food intake (“liking”) are processed by opioids.[236, 237] Those systems could be altered premorbidly as a vulnerability factor but also change in response to extremes of eating, which is an important part of our overall model of eating disorder pathophysiology.[238] Dopamine has a unique position. It is the only neurotransmitter that we have a mathematical understanding of neuronal function for, which can be used for computational modeling of brain function. Dopamine mediates reward learning[239] and has been implicated in the pathophysiology of taste and reward processing in eating disorders using the so-called prediction error model, which provides a very strong framework to study reward function in eating disorders.[240, 241, 242, 243, 244]
Although patients with bulimia nervosa (BN) are often unremarkable in general appearance and frequently have no signs of illness on physical examination, several characteristic findings may occur.
Physical findings may include the following:
Parotid hypertrophy. Reprinted with permission from Mandel, L and Siamak, A. Diagnosing bulimia nervosa with parotid gland swelling. J Am Dent Assoc 2004, Vol 135, No 5, 613-616.
![]() View Image | Parotid hypertrophy. Reprinted with permission [Mandel L, Siamak A. Diagnosing bulimia nervosa with parotid gland swelling. J Am Dent Assoc. 2004 May;.... |
See the list below:
Dental caries. Reprinted with permission from Wolcott, RB, Yager, J, Gordon, G. Dental sequelae to the binge-purge syndrome (bulimia): report of cases. JADA. 1984; 109:723-725.
![]() View Image | Dental caries. Reprinted with permission [Wolcott RB, Yager J, Gordon G. Dental sequelae to the binge-purge syndrome (bulimia): report of cases. J Am .... |
See the list below:
Russell sign. Reprinted with permission from Glorio R, et al. Prevalence of cutaneous manifestations in 200 patients with eating disorders. Int J Derm, 2000, 39(5), 348-353.
![]() View Image | Russell sign. Reprinted with permission [Glorio R, Allevato M, De Pablo A, et al. Prevalence of cutaneous manifestations in 200 patients with eating d.... |
Other cutaneous manifestations can include telogen effluvium (sudden, diffuse hair loss), acne, xerosis (dry skin), nail dystrophy (degeneration), and scarring resulting from cutting, burning, and other self-induced trauma.[68]
Other nonspecific but suggestive findings that may reflect the severity of the disease include bradycardia or tachycardia, hypothermia, and hypotension (often associated with dehydration). Edema, particularly of the feet (and less commonly the hands), is found more often among patients with a history of diuretic abuse, laxative abuse, or both or in patients with significant protein malnourishment causing hypoalbuminemia.
Some patients may be clinically obese, but morbid obesity is rare. Patients with BN who are overweight may have excessive fat folds that favor humidity and maceration with bacterial and fungal overgrowth, striae due to skin overextension, stasis pigmentation related to peripheral vascular disease, and plantar hyperkeratosis due to increased weight.[68]
A community-based household survey involving 52,095 adults in 19 countries found, after adjustment for pertinent comorbidities, that the rate of BN among the 2580 identified cases of adult-onset diabetes mellitus was twice that of non-diabetic individuals.[69]
A typical Mental Status Examination for a patient with BN is detailed below. (The formal Folstein Mini-Mental Status Examination [MMSE] is usually unnecessary in the evaluation of patients with BN because symptoms of dementia and delirium are not common in these patients.)
A national comorbidity study examined lifetime comorbidities of other psychiatric diagnoses in conjunction with bulimia nervosa (BN).[10] In general, the lifetime comorbidity of any psychiatric disorder is 94.5%.
The common co-occurrence of eating disorders with affective disorders suggests a possible relationship between them.[49] Major depressive disorder (MDD) is particularly common (approximately 50%) in this regard. Whether the association is causative (primary), secondary to the BN itself, or represents a common set of risk factors for BN and MDD is still unclear. Depressive symptoms can occur during pregnancy and postpartum in women with BN.[50] Bipolar II disorder also appears to be more common in patients with BN than in patients without eating disorders. The lifetime comorbidity for bipolar I or II is 17.7%.
Obsessive-compulsive disorder (OCD) is more common in persons with BN (17.4% lifetime comorbidity) than in controls. Panic disorder, social phobia, specific phobias, generalized anxiety disorder (GAD), and posttraumatic stress disorder (PTSD) significantly contribute to comorbidity. Lifetime comorbidities have been reported at approximately 17.4% for OCD, 16.2% for panic disorder, 41.3% for social phobia, 50.1% for specific phobias, 11.8% for GAD, and 45.4% for PTSD.
One study suggests that baseline clinical predictors such as female gender and family history of eating disorders might be specific to the later development of eating disorders in the context of childhood OCD.[51]
Some evidence suggested a relationship between disorders of substance abuse and dependence and BN, including alcohol dependence,[52] nicotine dependence,[53] and drug dependence.[54] For example, with regard to smoking in healthy controls, there were significantly more smokers among people with bulimia (lifetime OR = 2.165) and BED (lifetime OR = 1.792), but not for those with AN (lifetime OR = 0.927). Studies on caffeine intake are mixed.[55] Alcohol abuse or dependence has a lifetime comorbidity with BN of 33.7%, whereas illicit drug abuse or dependence has a lifetime comorbidity of 26%.
In a study of lifetime prevalence of impulse control disorders in patients with BN, compulsive buying, and intermittent explosive disorder were the most frequently reported disorders, at 17.6% and 13.2%, respectively. Higher than expected rates of kleptomania, pathological gambling, and trichotillomania have also been reported in patients with BN.[56]
ADHD may be associated with BN.[57] In one study of more than 2000 female inpatients treated for BN, 9% were also diagnosed with ADHD. The average rate of ADHD in the general population of young women is approximately 3.4%.[58] Lifetime comorbidity with ADHD and BN has been reported to be as high as 34.9%.
The role of sexual abuse in the development of eating disorders is controversial.[59] Some reports suggest a strong association, while others detect no increased association. Borderline personality disorder is found frequently.[60] These patients usually have histories of trauma and abuse and may represent a distinct subgroup. Pathologic narcissism[61] and identity impairment[62] may be present. Features of obsessive-compulsive personality disorder (OCPD), particularly perfectionism, may be increased among patients with BN.
BN is associated with increased risk of suicide attempts and suicidal ideations.[63] In one study, all-cause mortality rate for BN was 3.9%, higher than other studies reported in the past.[64] However, the standardized mortality ratio with respect to suicide was 6.51, a much higher than expected rate.
Studies suggest that patients with bulimia nervosa (BN) have increased rates of major depressive disorder, substance abuse, anxiety disorders, bipolar II disorder, and sexual abuse; these conditions should be considered and managed as necessary. Mortality and morbidity associated with depression (suicidal thoughts or self-injury) and poor impulse control (eg, substance abuse, sexually transmitted diseases, unintended pregnancy, accidental injuries) should always be anticipated and assessed. Patients with BN who are depressed and who have concurrent alcohol dependence are at exceptionally high risk of suicide, particularly those who overexercise. Studies have shown that in people with eating disorders, excessive exercise appears to be linked to an increased prevalence of acquired capability for suicide (ACS) and suicide attempts.[75, 76]
The all-cause mortality rate for BN per se is slightly lower than for anorexia nervosa (AN) (3.9% vs 4.0%, respectively).[64] Medical complications do arise and should be assessed carefully.
While the results of formal gastric emptying studies in patients with BN have yielded variable results (some suggesting delayed emptying time and others suggesting normal emptying time), acute gastric dilatation is a rare but concerning risk. This complication may result in gastric rupture, which may be fatal.
Among other rare potential complications are Mallory-Weiss tears of the esophagus, esophageal rupture, reflux esophagitis, and cardiomyopathies secondary to ipecac use.
Ipecac toxicity may be associated with skeletal myopathy, while chronic hypokalemia may also be associated with intestinal ileus, abdominal distension, exertional rhabdomyolysis, or both.
Hypokalemia-related distal renal tubulopathy is very rarely associated with BN.
Xerosis (dry skin) is a common finding in bulimia nervosa, which appears to be related to the chronic dehydration to which persons with BN are often prone.
Skin health usually requires an overall healthy nutritional status. Dermatological treatment is ordinarily topical.
Patients who chronically overuse and abuse laxatives risk chronic constipation, cathartic colon with pseudo-Hirschsprung syndrome, melanosis coli with increased risk for colon cancer, steatorrhea, and/or protein-losing enteropathy and metabolic consequences of hypophosphatemia and hypomagnesemia.
Other potential complications include osteopenia or osteoporosis, menstrual irregularity and infertility, and, less commonly, cognitive changes associated with dehydration and electrolyte and metabolic abnormalities.
A 21-year-old woman is brought into an outpatient clinic by her mother, who complains that her daughter has been demonstrating unusual eating patterns since she moved back home 6 months ago. Her mother observes her to eat large amounts of food, such as desserts, when she is alone, often finding food wrappers hidden in her daughter’s room. She is worried that her daughter may be engaging in vomiting after these episodes of heavy eating. She often isolates herself in the bathroom for 10–20 minutes after a large meal.
When the patient was asked about her eating habits, she admitted to a “loss of control.” She described feeling deep remorse when she eats more than she would like. Furthermore, she described feeling so laden with guilt about her eating binges that she purposefully induces vomiting at least once every other day. This act gives her tremendous relief. She admits that she is unhappy with her overall appearance and feels that she is “fat” and “out of shape.” She is preoccupied with her appearance and says that she compares herself to other women “all day long.” She also admits to feeling sad most days. She endorses experiencing occasional missed menstrual periods, low libido, low energy, and intermittent sore throat.
Historically, the patient has memories of a chaotic childhood. She is an only child whose parents fought often and finally divorced when she was 9 years old. The patient remembers the first time she induced vomiting at 10 years old, after she felt “too full after a large meal.” The mother describes her daughter as having few friends and as tending to isolate herself. However, the mother describes her as very bright; in fact, she was valedictorian of her high school.
On physical examination, the patient’s blood pressure is 90/60, heart rate is 100, and BMI is 19. Her oropharynx appears injected without areas of erosion, and multiple dental caries are seen. Bilateral parotid enlargement with minor tenderness is present. The patient is tachycardic and bowel sounds are hyperactive. The abdomen is soft, nontender, and nondistended. Skin turgor is poor.
On mental status examination, the patient presents as a young White woman with average body habitus and pale skin. She is meticulously dressed and groomed. She answers questions curtly, makes poor eye contact, and demonstrates mild foot tapping throughout the interview. Her mood is anxious, and her affect is mood congruent but restricted to negative emotionality. She is highly articulate. Thought process is linear and goal directed. Methodical about her statements, she often takes time to clarify what she “really means.”
Thought content displays themes of shame and self-reproach,[33] though it is unclear whether shame is a risk factor for the development of the eating problems or a consequence of these difficulties; while guilt may be present, its role is unclear. No active delusions or hallucinations are present. Her cognition is grossly intact. She denies suicidal thoughts, but sometimes wishes she was “invisible.” She has no violent or homicidal thoughts. Insight is limited regarding her ability to acknowledge her psychiatric illness. Her judgment is impaired considering her inability to recognize the potential negative health consequences of her eating behaviors.
Prior to entering the clinic, laboratory assessment obtained at the suggestion of the patient's primary care doctor reveals a serum potassium level of 3.8 Meq/L and serum amylase level of 140 Units/L.
Take-home points
This is important in detecting possible occult metabolic complications of bulimia. With significant vomiting, hypokalemic metabolic alkalosis is possible. Among patients with significant laxative abuse, normo-kalemic metabolic acidosis may occur. Hyponatremia, hypocalcemia, hypophosphatemia, and hypomagnesemia should be ruled out. Those who have significant intravascular depletion may have elevated blood urea nitrogen levels.
This is used to exclude anemia or other occult hematologic abnormalities. Leukopenia or pancytopenia should raise clinical suspicion that the person is significantly malnourished and may in fact meet criteria for typical or atypical AN.
Urine specific gravity may reflect the state of hydration. Some patients may water load in an attempt to gain some weight before their health care visit.
Comorbid substance abuse should be ruled out with a urine toxicology screen.[77] Some patients with suspected or reported purging may also use cannabis heavily, sometimes reporting that they use it to help with nausea. Persons with toxicology positive for cannabis and vomiting that they report is at times non-volitional may in fact have BN and comorbid cannabis hyperemesis syndrome.
Pregnancy status should always be obtained to rule out pregnancy in patients presenting with irregular periods. However, reports of amenorrhea or irregular menses, particularly if accompanied by weight loss, should raise a high index of clinical suspicion of diagnosis of AN (either typical or atypical AN) rather than BN.
Hyperamylasemia is found in up to 30% of persons with significant vomiting because of hypersecretion from the salivary glands. This may offer a rough measure of purging activity or suggest the presence of purging in patients who are suspected of purging but who deny doing it. However, serial amylase levels are not sufficiently sensitive to offer a useful clinical marker to follow the course of purging.
Imaging studies are rarely routinely indicated or ordered for uncomplicated or typical cases of bulimia nervosa (BN). For very complex prenstations that do not fulfill the typical criteria, an MRI of the brain may be indicated to rule out, for instance, a brain tumor.
Because of the potential for arrhythmias and cardiomyopathy as possible complications, an electrocardiogram should be performed in patients who are very thin, complaining of palpitations, or have other signs or symptoms suggestive of cardiovascular concern. Prolonged QTc, especially in the setting of hypokalemia, heighten the risk for cardiac decompensation in this population.[6]
Because of the potential for osteoporosis, a dual energy absorptiometry (DEXA) scan may be useful, particularly for patients with irregular menses, mood disorders, and/or who smoke cigarettes.
Routine neuropsychological testing is not indicated. When specific abnormalities are found in mental status testing, or when histories of learning impairment are present, neuropsychological testing may show decision-making abnormalities, as well as impairment in word recall, abstraction, attention, visuospatial functioning, and problem solving.[79, 80]
Most individuals with eating disorders do not receive treatment for their disorder. The "treatment gap" may involve individuals' attitudes about accessing various types of treatment, as well as perceived barriers to seeking treatment.[81] A study of people diagnosed or undiagnosed with eating disorders showed that undiagnosed individuals have more positive attitudes towards novel Internet- and smartphone-delivered treatments than diagnosed individuals. The strongest barriers to seeking treatment were fear of losing control, fear of change, and finding motivation to change. When adjusting for past experiences of treatment, eating disorder symptoms were positively associated with the strength of most treatment barriers.
Specifically, the most prominent perceived barriers to help-seeking were stigma and shame, denial of and failure to perceive the severity of the illness, and practical barriers (eg, cost of treatment); some cite lack of encouragement from others to seek help and lack of knowledge about help resources.[82] Therefore, programs targeting prevention and early intervention for eating disorders should focus on reducing stigma and shame, educating individuals about the severity of eating disorders, and increasing knowledge around help-seeking pathways for eating disorders. Mental health literacy, overall, may offer pathways into treatment. Not surprisingly, common themes reported by participants describe the process of recovery from anorexia nervosa (AN) include dealing with a fragmented sense of self, a turning point where insight and commitment to recovery is developed, and, in recovery, a reclamation of self through meaningful relationships, rebuilding identity, and self-acceptance.[83]
A range of psychotherapies and pharmacotherapies are efficacious for bulimia nervosa (BN),[33] as are a range of nutrition-based and self-help interventions. A meta-analysis of 79 RCTs included 19 different interventions, with primary outcome measures of abstinence from binge eating episodes, compensatory behaviors, and reduction of symptom severity; reduction of self-reported eating pathology and depression served as secondary outcome variables. Retrieved RCTs were meta-analyzed using fixed and random effects models. Overall, slightly reduced effects were obtained for self-help and moderate effects for pharmacotherapy. Similarly, psychotherapy yielded large-to-very-large effects in regard to secondary outcome variables, while moderate-to-large effects were observed for self-help, pharmacotherapy, and combined therapies. Followup analyses revealed the sustainability of psychotherapies in terms of large effects in primary outcome criteria, while these effects were moderate for self-help, pharmacotherapy, and combined therapies. The authors concluded that cognitive behavioral therapy (CBT) can be recommended as the best intervention for the initial treatment of BN.
Regarding binge eating disorder (BED), a meta-analysis of 45 studies found only moderate support for the efficacy of CBT and CBT-guided self-help, and modest support for interpersonal psychotherapy (IPT), selective serotonin reuptake inhibitors (SSRIs), and lisdexamfetamine in the treatment of adults with BED in terms of cessation of or reduction in the frequency of binge eating.[11]
Initial care for bulimia nervosa (BN) is usually provided in outpatient settings. Factors that may indicate a need for inpatient care include significant metabolic abnormalities, medical complications, risk of suicide, failed outpatient treatment, and inability to care for self.
Findings that suggest the need for inpatient treatment include:
The American Psychiatric Association (APA) recommends that adults with BN be treated with eating disorder–focused cognitive behavioral therapy (CBT) and that a serotonin reuptake inhibitor (eg, 60 mg fluoxetine daily) may also be prescribed, either initially or if there is minimal or no response to psychotherapy alone by 6 weeks of treatment.[120]
The APA suggests that adolescents and emerging adults with BN who have an involved caregiver be treated with eating disorder-focused family-based treatment (FBT).[120] The American Academy of Child and Adolescent Psychiatry also supports the use of FBT in youth but also recognizes that CBT may be equally effective especially on longer follow up.[140]
BN is best managed using an interdisciplinary approach. Care providers who should be involved include the primary care provider, psychiatrist, psychotherapist, and nutritionist/dietitian. If the psychiatrist is not skilled in this area, involvement of a psychotherapist with expertise in the management of eating disorders is strongly recommended. Dietary review and nutritional rehabilitation counseling should be provided by a nutritionist/registered dietitian. Dental care merits attention. Depending on complications, those with BN may also require the services of other specialists. The goals of treatment are as follows:[84]
Most authorities agree that patients with BN who have benefited from CBT are likely to benefit from a program of relapse prevention as well, lasting usually 1–2 years. Since simply telling patients with BN who have achieved abstinence following a course of CBT to return for additional sessions if they fear relapse has been ineffective for preventing relapse, planned visits or regularly scheduled phone calls might be used.[85]
Treatment for bulimia nervosa (BN) should be comprehensive and generally requires an interdisciplinary approach with many of the components described below.[8]
A systematic meta-review of meta-analyses and network meta-analyses[138] indicated that, in adults with BN, individual cognitive behavioral therapy (CBT) had the broadest efficacy versus active control; also, antidepressants outperformed active control.
BN was investigated in one network meta-analysis and 19 meta-analyses; two (10.0%) reported data on adolescents, six (30.0%) on adults. For adolescents, compared to active control, in outpatients, individual family-based therapy improved global course of the disease at follow-up (k = 2, n = 37, OR = 2.22, 95% CI = 1.02–4.82, AMSTAR/ Quality = 9/H), whereas in mixed settings CBT for eating disorders did not differ from active control.
In adults, 42.9% of the interventions proved beneficial and 7.1% had both beneficial and detrimental effects. In outpatients, only group CBT for eating disorders was covered, compared to wait list-no treatment, with efficacy regarding eating disorder behaviors (k = 4, n = 51, SMD = 0.56, 95% CI = 0.15–0.96, AMSTAR/Quality = 9/H) and global course of the disease (k = 2, n = 31, RR = 1.29, 95% CI = 1.04–1.61, AMSTAR/Quality = 9/H). In mixed settings, compared to wait list-no treatment, individual CBT for eating disorders and individual psychotherapy were both beneficial for eating disorder behaviors. Global course of the disease was ameliorated also by group behavioral therapy (k = na, n = 207, OR = 28.7, 95% CI = 3.11–455, AMSTAR/Quality = 8/H), group CBT for eating disorders (k = na, n = 245, OR = 7.67, 95% CI = 1.51–55.7, AMSTAR/Quality = 8/H), and self-help CBT for eating disorders (k = na, n = 302, OR = 3.49, 95% CI = 1.20–11.2, AMSTAR/Quality = 8/H; k = na, n = 392, OR = 3.81, 95% CI = 1.51–10.9, AMSTAR/Quality = 8/H).
Compared to active control, individual CBT for eating disorders had the broadest efficacy, namely for eating disorder behaviors (k = 25, n = na, g = 0.23, 95% CI = 0.06–0.37, AMSTAR/Quality = 5/M), EDP (k = 16, n = na, g = 0.27, 95% CI = 0.08–0.45, AMSTAR/Quality = 6/M), and global course of the disease (k = 7, n = 484, OR = 1.20, 95% C = 1.03–1.41, AMSTAR/Quality = 5/M; k = 14, n = na, OR=1.70, 95% CI = 1.21–2.38, AMSTAR/Quality = 6/M).
Individual psychotherapy was worse than control for global course of the disease (k = 6, n = 257, RR = 0.83, 95% CI = 0.69–0.98, AMSTAR/Quality = 8/H).
Antidepressants were beneficial for eating disorder behaviors (k = 3, n = 89, RR = 2.08, 95% CI = 1.09–25.0, AMSTAR/Quality = 8/H) and general psychiatric symptoms (k = 3, n = 117, RR = 2.13, 95% CI = 1.09–100, AMSTAR/ Quality = 8/H).
Individual therapies
Two systemic reviews have evaluated psychotherapy, psychopharmacology, and other treatment interventions. The first focused on 21 studies in the United Kingdom and the outcome analyzed was full remission at the end of treatment.[86] The study included 12 treatments, including wait list, and found individual CBT was most effective in achieving remission at the end of treatment compared with wait list (OR 3.89, 95% CrI 1.19–14.02), followed by guided cognitive behavioural self-help (OR 3.81, 95% CrI 1.51–10.90). The other meta-analysis of 79 RCTs with 19 interventions revealed moderate-to-large intervention effects for psychotherapy, primarily CBT, and substantial sustainability of psychotherapies compared to self-help, pharmacotherapy, and combined treatments.[33]
CBT is an evidence-based, effective treatment for BN. Behavioral approaches to avoiding undesirable eating habits are used, including diary keeping; behavioral analyses of the antecedents, behaviors, and consequences (so-called ABCs) associated with binge eating and purging episodes; and exposure to food paired with progressive response prevention regarding binge eating and purging. Distorted or maladaptive thoughts regarding weight and shape are identified, examined, and addressed, and other dysfunctional irrational beliefs are explored and confronted to allow better understanding, enhanced self-control, and improved body image. The cognitive component of CBT appears to be an essential active ingredient for change, as behavioral interventions alone are often not as effective.[84]
CBT is indicated as first-line treatment. Patients with other comorbid personality disorders, specifically borderline personality disorder, may not respond to conventional CBT to the same degree as those without these comorbidities. The more complex patients have better experiences with “enhanced” versions of CBT (specifically CBT-E) consisting of CBT augmented with techniques taken from interpersonal psychotherapy and other forms of psychotherapy.[87] CBT may help to reduce binge eating and purging behavior but is generally not effective alone for producing weight loss.[88]
A systematic review of CBT-enhanced (CBT-E) RCTs and open trials for BN, binge eating disorder (BED), and transdiagnostic samples found 7 effectiveness studies (5 randomized and 2 open trials).[89] Substantial differences in posttreatment remission rates were found (range: 22.2%–67.6%) due, in part, to differences in samples and operationalization of clinical significant change, though more studies on differential effects and working mechanisms are required to establish the specificity of CBT-E.
Interpersonal psychotherapy
Interpersonal psychotherapy (IPT) addresses specific issues in the interpersonal arena that create the context for and stimulate dynamic tensions that spur the patient's symptoms; these generally encompass such processes as grief, role transitions, role conflicts or disputes, and interpersonal deficits. Brief focused therapy in these areas can be effective in producing improvements in mood disturbance and low self-esteem, which may trigger and maintain the symptoms of BN. The efficacy of IPT is roughly similar to CBT in reducing binge eating, but it may be somewhat less effective in curbing purging.
A systematic review of 15 studies (RCTs and long-term, follow-up studies derived from the RCTs) revealed 6 main findings:
IPT is a reasonable, cost-effective alternative to CBT for the overall eating disorders spectrum.
Nutritional rehabilitation counseling
A structured meal plan provides a concrete means to help reduce episodes of dietary restriction and, consequently, urges to binge and purge. Adequate nutritional intake can prevent craving and promote satiety. Assessing nutritional intake for all patients, even those with a normal body weight (and normal BMI), is important because normal weight per se does not ensure appropriate nutritional intake or normal body composition. Nutritional counseling may not only help reduce food restriction, but may also help patients increase the variety of foods eaten and promote healthy but not compulsive exercise patterns.
Family therapy
Family therapy explores family attitudes and dynamics, dysfunctional relationships, communication and behavior patterns, and other factors that may precipitate or perpetuate abnormal eating behaviors in the family setting. This perspective often views eating difficulties as a form of communication within a family. Family therapy should be considered especially for adolescent patients still living with their parents or older patients with ongoing conflicted interactions with parents. A specific form of family intervention, the so-called Maudsley model, where parents are authorized to take an active and specified role in helping their child to eat, has been shown to be effective for treating AN in adolescents.
For adolescents, a randomized controlled trial showed family-based treatment to be more effective for achieving abstinence than CBT at the end of 6 months of treatment. However, little difference was seen between these two treatments at the end of one year.[91]
Individual therapy
Psychodynamic psychotherapy: Some patients, particularly those with concurrent developmental and personality pathology or other co-occurring disorders, require lengthy individual treatment. Clinical reports suggest that psychodynamic and at times psychoanalytic approaches in individual or group format may help to improve overall coping once bingeing and purging improve. Supportive-expressive psychotherapy (SEP) in individual or group therapy formats may be helpful for patients with BN.
Couples therapy
Patients with marital discord may benefit from couples therapy. Although research concerning marital and cohabitation relationships in patients with BN is limited, many patients with BN are thought to experience problematic relationships with impaired intimacy, including sexuality and suboptimal communication skills. Self-consciousness and self-silencing associated with sexual activity, and anxious attachment may be associated with bulimic symptoms.[92]
Group therapy
Dialectical behavior therapy and integrative cognitive-affective therapy may have an emerging role in treatment, especially for patients with severe emotional dysregulation and impulsivity.[93]
Virtual reality
Virtual reality (VR) is a technology with a variety of uses including clinical research, education/training, and assessment and treatment of several medical and psychological conditions. A colloquial definition is that it is a human–computer interface that allows the user to interact with and become immersed in a computer-generated environment, which produces the feeling of “being there.” A review of 19 studies on the use of VR in BN and BED found 9 studies on assessment and 10 on treatment.[94] Though the research is at an early stage, the use of VR in the assessment of those conditions showed some promise in identifying: 1) how those patients experienced their body image; and 2) environments or specific kinds of foods that may trigger binge–purging cycle. Some studies using VR-based environments associated to CBT showed their potential utility in improving motivation for change, self-esteem, body image disturbances, and in reducing binge eating and purging behavior.
Self-help and support groups
Support groups and 12-step programs such as Overeaters Anonymous may be helpful as adjuncts in initial treatment and for subsequent relapse prevention, but they are not recommended as the sole initial treatment approach for bulimia nervosa. In the most recent update of the Self-help and guided self-help for eating disorders in the Cochrane Database of Systematic Reviews,[95] efficacies of pure self-help (PSH) and guided self-help (GSH) were mixed. PSH/GSH did not significantly differ from waiting list in abstinence from binging or purging, although PSH/GSH produced greater improvement on other eating disorder symptoms, psychiatric symptomatology, and interpersonal functioning, but not depression. In addition to face-to-face interventions, self-help support may be available through books and online. See the APA's complete list of Self-Help Books and Internet Resources in the Patient Education section.
Bright light therapy has been shown to reduce binge frequency in several controlled trials of patients whose binge eating follows a seasonal pattern (akin to seasonal affective disorder) and may be used as an adjunct when CBT and antidepressant therapy have not been effective in reducing bingeing symptoms in such patients.[96]
One study provided some support for guided imagery compared to journaling. However, long-term maintenance of treatment effects is unknown.[97]
Recent research has shown Web-based CBT to improve psychopathology, decrease body dissatisfaction, and benefit other problems related to eating disorders. Overall, technology-based interventions (using computers, the Internet, and mobile resources) have been efficacious both in preventing and treating eating disorders.[98, 99]
Fluoxetine (Prozac)
In clinical trials, initial dose 20 mg/d with advance over 1–2 weeks to 60 mg/d in the morning as tolerated. Some patients may need to begin at a lower dose if side effects are intolerable. A maximum dose of 80 mg/d may be used in some cases.
Two large multicenter studies indicated reduced binge eating and purging behaviors with fluoxetine over placebo but also greater side effects such as insomnia, nausea, or tremor.[269, 269] Higher fluoxetine dose provided a stronger therapeutic effect on pathologic eating attitudes, carbohydrate cravings, and depression, and the current recommendation is to quickly titrate the medication to 60 mg in adults with BN. However, a high dropout rate was concerning, with 45% in the 16-week study not completing treatment. Follow-up studies further supported the combination of medication with CBT.[270, 271] On long-term follow-up, only 17% of the fluoxetine and 8% of the placebo-treated patients completed the study, suggesting that fluoxetine’s effectiveness is limited, especially over time.[272]
Antidepressants
Tricyclic antidepressant medications tried in BN indicated improvements in mood and ED behaviors,[245, 246, 247] but group therapy effects were stronger than active drugs, and added drugs were not superior to psychotherapy alone for core eating disorder behaviors.[248] For instance, imipramine plus group therapy resulted in less anxiety and depression than group therapy alone, highlighting that the pathophysiology of eating disorder behaviors differs from depression and anxiety, requiring different treatments. Desipramine versus placebo resulted in reduced eating disorder behaviors, but relapse was high, as was the dropout rate, up to 50%,[249, 250, 251, 252, 253] while the tetracyclic antidepressant mianserin did not lead to significant improvements over placebo.[254]
Monoamine oxidase inhibitors (MAOIs) reduced binge eating frequency in BN; however, the dropout rate across studies was high.[255, 256, 257, 258, 259, 260]
The norepinephrine/dopamine-reuptake inhibitor bupropion showed promise in treating BN; however, 7% of active drug probands suffered grand mal seizures, and bupropion is contraindicated in eating disorders associated with purging behaviors.[261]
The serotonin reuptake and histamine and alpha-1-adrenergic receptor blocker trazodone, compared to placebo, reduced binge eating and purging frequency in BN and improved self-control in one study, but no further investigations were conducted.[262]
Studies using the SSRI fluvoxamine were mixed, dropout was high, up to 75%, and potential side effects were concerning, including elevated seizure frequency and liver function test abnormalities.[263, 264, 265, 266, 267] The strongest evidence for pharmacological efficacy for treating BN exists for the SSRI fluoxetine, which has FDA approval.
Clinicians must be aware of the black box warnings relating to antidepressants and other medications to discuss the potential benefits and risks as part of the consent process with patients and families if such medications are to be prescribed. See the statement on Antidepressant Use in Children, Adolescents, and Adults by the Food and Drug Administration.
Mood stabilizers
BN is associated with mood fluctuations, which may trigger binge eating episodes; however, the mood stabilizer lithium was not superior to placebo.[273] Additionally, as lithium, especially with long-term usage, can cause kidney disease, it should be used with caution in patients with active purging. Purging can cause dehydration and unwanted rise in lithium levels, which increases the risk of possibly irreversible nephrotoxicity from lithium usage. The antiepileptic topiramate is often used off-label for mood stabilization, reduced binge eating and purging frequency, and depression and anxiety in BN[274, 275] However, the side effect profile of topiramate can be potentially quite problematic for persons trying to recover from bulimia, as it can cause weight loss and cognitive blunting, both of which could impede recovery for persons with bulimia nervosa.
Other agents
Ondansetron, a competitive serotonin 3 receptor antagonist used to treat nausea and vomiting, showed improved binge eating and purging in a small sample,[276] but no follow-up studies support ondansetron's use in BN.
Case reports indicate that methylphenidate may be helpful for patients with BN and concurrent ADHD.[120] Stimulants should be used with caution in patients with eating disorders, including patients with BN, due to risks of weight loss and potentially increased risk of sudden cardiac arrthymia in a population already at high risk of cardiac events (mostly due to electrolyte derangements secondary to purging).
Trials of traditional and nontraditional medication treatments have to be weighed in terms of potential for drug interactions, the medical complications of BN, and the medical comorbidities of BN.
Patients with BN often benefit more from combinations of psychotherapy and pharmacotherapy than from either treatment alone, particularly in the presence of a comorbid depressive disorder, which is seen in the majority of cases. For uncomplicated BN, CBT alone is superior to pharmacotherapy alone.[111]
Cognitive behavioral therapy (CBT) is the single most well-studied and effective treatment for bulimia nervosa (BN). Some studies have reported that the combination of antidepressant therapy and CBT results in the highest remission rates. This combination is recommended initially when qualified CBT therapists are available. In addition, when CBT alone does not result in a substantial reduction in symptoms after 10 sessions, addition of fluoxetine is ordinarily recommended.[120] However, a major study found that when excellent manual-based CBT is administered, the addition of fluoxetine may not offer additional benefit. How best to treat individuals who do not respond to CBT and/or antidepressant medications remains an unsettled question.[97]
Limited evidence supports the use of fluoxetine for relapse prevention, but substantial rates of relapse occur even with treatment. The optimal duration of treatment and the optimal strategies for maintaining treatment gains are unknown. In the absence of adequate data, most clinicians recommend continuing antidepressant therapy for a minimum of 9 months and probably for at least 1 year in most patients.[8] Other medications have not been studied long term in BN.
With CBT and maintenance treatment, as many as 50% of patients with BN are asymptomatic at follow-up 2–10 years after completing treatment. Rates of persistent long-term improvement following other forms of psychotherapy (eg, IPT, supportive-expressive psychotherapy) are unknown.
In an update of Psychotherapy for bulimia nervosa and binging in the Cochrane Database of Systematic Reviews, CBT, particularly CBT-BN (a specific modification of CBT to address bulimia nervosa), noted efficacy in decreasing binge eating. However, these conclusions are limited by the fact that the clinical trials were highly variable and small sample sizes. Long-term IPT was also demonstrated to be efficacious. Self-help, alongside highly structured CBT, appeared to be promising. However, exposure and response prevention did not appear to enhance the efficacy of CBT. Psychotherapy alone was unlikely to change body weight.[112]
According to the most recent update of Antidepressants versus psychological treatments and their combination for bulimia nervosa in the Cochrane Database of Systematic Reviews, combination treatments of medications plus psychotherapy were superior to psychotherapy alone. Psychotherapy appeared to be more acceptable to patients. When antidepressants were combined with psychological treatments, acceptability of the latter was significantly reduced.[113]
Technology-based interventions, such as Internet prevention programs, Internet-assisted CBT, online consulting, and text messaging have shown promise in assisting in the treatment of eating disorders.[114, 115, 116]
Major medical treatment requiring surgical intervention is rare in bulimia nervosa (BN), but medical care providers should be familiar with potential serious complications.
Patients may develop an acute gastric obstruction and/or gastric dilatation[117] (rarely resulting in gastric perforation leading to acute peritonitis), which presents with severe, continuous projectile vomiting that occurs soon after any oral intake. This possibility should be considered in individuals with known BN who present complaining of uncontrollable vomiting. When the potential for gastric dilatation, outlet obstruction, or both is of concern, an urgent surgical consultation is indicated.
Emergency surgical review is also required if symptoms suggestive of esophageal tear (Mallory-Weiss syndrome) develop or in case of esophageal rupture, which can precipitate acute mediastinitis. (See images below). For more information, see Medscape Reference articles Mallory-Weiss Syndrome and Esophageal Rupture.
This chest radiograph demonstrates pneumomediastinum, which can occur in association with esophageal rupture from forceful vomiting.
![]() View Image | This chest radiograph demonstrates pneumomediastinum, which can occur in association with esophageal rupture from forceful vomiting. |
Water-soluble contrast esophagram from a patient with esophageal perforation after esophageal dilation shows contrast leak (arrowheads) and normal esophageal lumen (arrows).
![]() View Image | Water-soluble contrast esophagram from a patient with esophageal perforation after esophageal dilation shows contrast leak (arrowheads) and normal eso.... |
Mallory-Weiss tear. Typical longitudinal mucosal tear with overlying fibrinous exudate extending from the distal esophagus to the gastric cardia. Courtesy of C.J. Gostout, MD.
![]() View Image | Mallory-Weiss tear. Typical longitudinal mucosal tear with overlying fibrinous exudate extending from the distal esophagus to the gastric cardia. Cour.... |
These conditions are surgical emergencies and, although uncommon, are occasional causes of mortality related to BN.
Dentists and dental hygienists sometimes play a unique role in opening dialogues with patients about eating disorders. They can help with early recognition and refer patients for specialist-level eating disorders care. Similarly, dental professionals can make important contributions to recovery and long-term treatment of these patients.[118]
As described above regarding the role of the registered dietician, patient education regarding healthy, well-balanced diets, exercise, and long-term maintenance of a healthy weight is important in bulimia nervosa (BN) and may help reduce the risk of relapse or chronicity. Key here is often that patients follow a strict meal plan that helps re-habituate to regular meals. Whether a specific nutritional composition is beneficial for BN is currently being researched.
In depictions of persons with eating disorders in the media and in popular culture, emphasis is often placed on the potential etiological role of body image insecurities and a desire for thinness present in both bulimia nervosa (BN) and anorexia nervosa (AN). However, in many developed countries, thinness is valued, especially among women and girls, but the vast majority of women and girls do not develop AN or BN. It should be noted, however, that some studies have found that a culture that places a premium on having a thin body habitus may increase the risk of subthreshold disordered eating behaviors. Furthermore, not all persons with AN or BN will express the primary driver of their eating disorder behaviors as a desire to be thin.
Awareness of the cultural and social forces and education for both children and their parents regarding the attitudes and behaviors that foster eating disorders may reduce the prevalence of these syndromes. Opportunities for this kind of intervention abound in primary care, athletic, and educational settings. School-based programs that emphasize health, fitness, and a range of physical and psychological competences have shown promise in being able to reduce the development of eating disorder-associated attitudes in vulnerable school-age populations.[119]
A dissonance-based thin-ideal internalization reduction program has been studied in the past as an eating disorder prevention program. That prevention intervention led to greater decreases in thin-ideal internalization and weight than did a control condition. Dissonance program participants showed significantly greater decreases in thin-ideal internalization, body dissatisfaction, negative affect, eating disorder symptoms, and psychosocial impairment and lower risk for eating pathology onset through 2- to 3-year follow-up than did assessment-only controls.
Bulimia nervosa (BN) is an often-chronic disorder with relapse or cross over to anorexia nervosa (AN) or binge eating disorder (BED) and long-term monitoring such as intermittent check-in with a health professional is frequently indicated.
The American Psychiatric Association (APA) published guidelines for the treatment of eating disorders in February 2023.[120, 121] The guidelines recommend that screening for the presence of an eating disorder should be part of an initial psychiatric evaluation. The initial evaluation of a patient with a possible eating disorder should include assessment of multiple factors, including but not limited to, patient’s history of height and weight, eating-related behaviors, food repertoire, weight control behaviors, and family history. The evaluation should also identify co-occurring health conditions and psychiatric disorders.
Adults with bulimia nervosa (BN) should be treated with eating disorder–focused cognitive behavioral therapy (CBT) and prescribed a serotonin reuptake inhibitor.
Canadian practice guidelines for treating children and adolescents with eating disorders were published in 2020.[122] Recommendations include the following:
The British National Institute for Health and Care Excellence (NICE) guidance for eating disorders recommends the following:
Anorexia Nervosa
AN and BN
Bulimia Nervosa
Restriction of energy intake relative to requirement, leading to significantly low body weight in the context of age, sex, developmental trajectory, and physical health Recurrent episodes of binge eating: a) eating in a discrete period of time, within a 2-hour period, an amount that is definitely larger than what most would eat in that time period; b) a sense of lack of control over eating Intense fear of gaining weight or of becoming fat, or persistent behavior that interferes with weight gain, even though at a significantly low weight Recurrent inappropriate compensatory behaviors in order to prevent weight gain, such as self-induced vomiting; misuse of laxatives, diuretics, or other medications; fasting; or excessive exercise Disturbance 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 The binge eating and inappropriate compensatory behaviors both occur, on average, at least once a week for 3 months Two types: Restrictive and Purge/Binge-Eating type The disturbance does not occur exclusively during episodes of AN Body Dysmorphic Disorder BDD and BN BN Preoccupation with one or more perceived defects or flaws in physical appearance that are not observable or appear slight to others Disturbance in which one's body weight or shape is experienced, undue influence of body weight or shape on self-evaluation Recurrent episodes of binge eating: a) eating in a discrete period of time, within a 2-hour period, an amount that is definitely larger than what most would eat in that time period; b) a sense of lack of control over eating At some point during the course of the disorder, the individual has performed repetitive behaviors or mental acts in response to the appearance concerns The preoccupation causes significant distress or impairment in social, occupational, or other important areas of functioning Recurrent inappropriate compensatory behaviors in order to prevent weight gain, such as self-induced vomiting; misuse of laxatives, diuretics, or other medications; fasting; or excessive exercise The appearance preoccupation is not better explained by concerns with body fat or weight in an individual whose symptoms meet diagnostic criteria for an eating disorder The binge eating and inappropriate compensatory behaviors both occur, on average, at least once a week for 3 months Differential Diagnosis of Bulimia Nervosa[1]