AUTHOR AND EDITOR INFORMATION

Section 1 of 10  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• Follow-up
• Miscellaneous
• References

INTRODUCTION

Section 2 of 10  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• Follow-up
• Miscellaneous
• References

Background

For thousands of years, humans have engaged in occasional eating binges when adequate food supplies are available. The practice of vomiting after overeating also dates back thousands of years. The examination and definition of abnormal eating patterns as eating disorders did not occur until roughly 20 years ago. Bulimia nervosa (BN) was first described in 1979 as a disorder that involves binge eating followed by inappropriate behavior to avoid weight gain. Criteria for the diagnosis of BN did not emerge until provided by the American Psychiatric Association in 1980.

Criteria for the diagnosis of bulimia include preoccupation with eating and overeating large amounts of food in short periods, also described as binge eating. This behavior is then followed by inappropriate behavior to avoid weight gain, most notably, self-induced vomiting. Other methods of avoiding weight gain include laxative and diuretic abuse and excessive exercise. BN is a disease with a highly focused patient population; it is predominantly found in women and is virtually nonexistent in nonindustrialized countries.

Pathophysiology

BN is a disease that most likely emerges from a complex integration of many factors. These factors may be psychological, cultural, environmental, and societal. Many proposed associated factors are involved in the development of bulimia. These factors can include chemical imbalances in neurotransmitters, such as serotonin or pancreatic polypeptides (eg, pancreatic peptide YY [PYY]). Psychological and psychiatric problems are also thought to contribute to the development of bulimia. Another contributing factor is family problems. Participation in extracurricular activities that emphasize body shape and image has also been linked to the development of bulimia.

The binge and purge cycle characteristic of bulimia affects multiple organ systems. The GI system can be affected by the overeating associated with binge episodes. This overeating can stretch the stomach or delay gastric emptying. Purging can induce esophagitis or esophageal rupture due to vomiting. Pancreatitis can also occur. Electrolyte abnormalities can include hypokalemia and hypochloremia. Cardiovascular abnormalities can lead to arrhythmias, arrest, cardiac rupture, or pneumomediastinum. The pulmonary system can be damaged by aspiration of gastric contents upon vomiting. Renal function impairment is also possible.

Frequency

United States

Lifetime prevalence among women is 1-3%, and a comparable percentage of women have less severe variants of the disorder. Lifetime prevalence among men is 0.1%.

International

Although no concrete data are available, bulimia is a disease that is highly culturally dependent. It is found solely in societies in which a high cultural value is placed on slimness and is virtually nonexistent in nonindustrialized countries.

Mortality/Morbidity

• Death is a relatively uncommon outcome for bulimia. Approximately 0-3% of women with the disease eventually die from complications of the disease; however, these numbers may be underestimated due to low ascertainment rates and short follow-up periods.
• The leading cause of death among patients with eating disorders is suicide, which is more common in patients with BN than in those with anorexia nervosa (AN). Factors most strongly associated with suicide attempt or suicidal ideation in patients with eating disorders include concurrent drug use, alcohol use, and tobacco use.¹ Suicide risk should be carefully monitored in patients with eating disorders who also have these risk factors.
• One third of patients who present for treatment of BN have past histories of AN.
• Bulimia has many complications (see Complications).

Race

Bulimia has traditionally been thought of as a disease that predominantly affects whites. The low incidence of eating disorders among nonwhites has been attributed to differences among ethnic groups in ideal body image. Studies have shown that black women are less likely to develop eating disorders and tend to express more satisfaction with their bodies than white women of similar weight; however, other studies suggest that the incidence of bulimia among minority groups is higher than previously thought. Studies suggest that patients from higher socioeconomic groups are more likely to seek treatment, making the incidence within these groups appear to be higher.

Some population studies suggest an equal incidence of bulimia in blacks and whites. Overall, strong circumstantial evidence suggests that cultural factors play large roles in eating disorder development. Most cases of BN originate in industrialized countries. In general, industrialized countries are places where food is plentiful and a preoccupation with thinness in women is present.

Sex

Bulimia primarily occurs in young women. Males comprise only 2-8% of all bulimia cases.

Age

Bulimia is most common in adolescents and young adults. Median age of onset is 18 years.

CLINICAL

Section 3 of 10  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• Follow-up
• Miscellaneous
• References

History

Obtaining a thorough history is essential in any patient in whom bulimia is suggested.

• Patients often deny the problem; however, thorough and careful questioning may reveal clues that the patient has bulimia.
• Often, patients have a history of dieting attempts and may admit to feeling fat even when they appear thin.
• Patients often state that their self-esteem is linked closely to their body weight or shape.
• The patient may have a history of using diet pills, laxatives, ipecac, or thyroid medication to lose weight.
• Patients may become vegetarians.
• Diabetic patients may withhold insulin.
• Patients who admit to purging behavior often describe a history of uncontrolled eating binges at least twice weekly. 
• • During these binges, large amounts of food are consumed in private. Some patients plan ahead for binges by secretly hoarding food.
  • Patients may describe feeling a loss of control when the binge begins, then a period of frenzied and rapid eating.
  • The binge is followed by inappropriate compensatory behavior, usually self-induced vomiting.

Physical

The diagnosis of bulimia is not conditional on physical findings. Physical findings may not be present in all patients. Patients may have some findings, all findings, or none at all. The following physical findings are associated with bulimia:

• General  
• • Normal body weight, but may be increased or decreased
  • Often frequent fluctuations in weight
  • Loss of subcutaneous fat
• Vital signs 
• • Temperature - Hypothermia
  • Blood pressure - Hypotension
• Head, ears, eyes, nose, and throat
• • Dental erosion - Decalcification of dental surfaces exposed to vomitus (Amalgams and fillings are relatively resistant to acid and often project above the surface of the teeth.)
  • Palatal trauma
  • Painless enlargement of parotid glands
• Cardiovascular - Bradycardia
• Abdomen - Frequent complaints of diffuse pain upon palpation
• Extremities 
• • Metacarpal phalangeal bruises, calluses, scarring, abrasions (Russell sign)
  • Edema possible if patient abuses laxatives or diuretics
  • Proximal muscle weakness if patient abuses ipecac
  • Waddling gait if patient abuses ipecac

Causes

Bulimia nervosa (BN) is a complex disease that most likely emerges from an integration of physiological, psychological, and environmental factors. Currently, no defined single cause of BN is recognized. Several factors are believed to play a strong role in the development of bulimia.

• Chemical: A few hypotheses suggest specific chemical abnormalities in the body are associated with bulimia.
• • Serotonin is a neurotransmitter with broad functions within the body. Among these functions, serotonin is involved in the development of satiety. Increased levels of serotonin are associated with decreased food intake. Serotonin is believed to increase postprandial satiety rather than directly decrease appetite.
  • • One hypothesis of the development of bulimia involves abnormalities in serotonergic function. Some patients with bulimia have been found to have low serotonin levels. Because serotonin is involved in the development of satiety, these disturbances may contribute to the persistence of binge eating. 
    • A potential hypothesis is that an impaired serotonergic response may contribute to the blunted satiety and prolonged periods of rapid food ingestion present in BN. Dieting has also been associated with altered serotonin function, more markedly in women than in men. Dieting is often a precursor to the development of bulimia; however, not all women who diet develop bulimia. This hypothesis is not thought to provide a sufficient sole explanation for the development of bulimia.
  • Another suggested pathophysiology involves increased levels of peptides involved in mediating appetite. Increased levels of a pancreatic polypeptide PYY, a peptide known to increase appetite, have been found in some patients with bulimia after a period of eating stability. This would suggest that these patients have a higher level of appetite, even when given a normal diet.
• Psychiatric
• • Premorbid psychiatric disorders are often associated with development of bulimia. These can include affective disorders, anxiety disorders, and substance abuse.
  • Many patients with bulimia have concomitant depression.
• Psychological and environmental
• • The strongest risk factor in the development of bulimia is history of dieting. Many patients report that their eating binges began in the context of or immediately following a diet. Many patients continue to restrict their caloric intake even when not binge eating.
  • Strong circumstantial evidence suggests that cultural factors play a large role in eating disorder development. Most cases of BN originate in industrialized countries where food is plentiful and a preoccupation with thinness in women is present.
  • Obesity is another risk factor for bulimia.
• Family
• • Family problems are also associated with the development of bulimia.
  • A history of sexual abuse has been associated in some literature as a risk factor for development of bulimia.
  • A family history of eating disorder increases a child's risk of developing an eating disorder to 2-20 times that of the general population.
• Interests and activities
• • Certain athletes and groups are thought to be more prone to development of bulimia. Specifically, these include ballet dancers, models, cheerleaders, runners, gymnasts, weight lifters, body builders, jockeys, divers, wrestlers, figure skaters, and field hockey players. Persons in these particular sports and activities often place a high value upon thinness or maintaining a particular weight.
  • The bodies of participants in these activities are often on display in front of crowds or judged in terms of body shape and weight. These high-pressure situations and preoccupation with weight can place teens at risk for eating disorders.

DIFFERENTIALS

Section 4 of 10  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• Follow-up
• Miscellaneous
• References

Other Problems to be Considered

Body Dysmorphic Disorder
Psychogenic vomiting
Klüver-Bucy–like syndromes
Kleine-Levin syndrome
Major depressive disorder
Anorexia, binge-eating and purging type
Hypothalamic brain tumor
Epileptic equivalent seizures
Medications - Lithium, tricyclic antidepressants, neuroleptics, insulin, opiates
Increased intracranial pressure
CNS tumors
Anxiety disorder
Substance abuse
Cluster B personality disorder

WORKUP

Section 5 of 10  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• Follow-up
• Miscellaneous
• References

Lab Studies

• No single diagnostic laboratory study is indicated for bulimia.
• A chemistry panel may be obtained if dehydration or electrolyte imbalances are suspected. Common findings include the following:
• • Hypokalemia - Secondary to vomiting
  • Increased BUN levels
  • Decreased magnesium
  • Decreased chloride
• Patients with bulimia nervosa (BN) may exhibit the following abnormal endocrine findings:
• • Increased basal serum prolactin
  • Positive dexamethasone suppression test result
• If ipecac use is suggested, the following lab studies may be indicated:
• • Stool and urine for emetine (byproduct of ipecac)
  • Cardiac assessment - Muscle enzyme values, lipid levels, magnesium, zinc, electromyography
• Drug screen is indicated for patients with possible drug use.
• GI tests may reveal the following:
• • Increased amylase secondary to vomiting
  • Liver function test results usually normal

Imaging Studies

• Imaging studies are not indicated; however, CT scan and MRI of head have demonstrated pseudoatrophy in some patients with BN, suggesting problems with malnutrition. The degree of atrophy with BN is less than that observed with anorexia.

Other Tests

• ECG should be considered in the following conditions:
• • Suspected ipecac abuse
  • Hypokalemia
  • Patient has experienced symptoms and signs of arrhythmias
• Gastric motility studies should be considered in the following conditions:
• • Prolonged history of bulimia
  • History of constipation
  • Other unexplained abdominal pain

Procedures

• No diagnostic procedures are indicated.

TREATMENT

Section 6 of 10  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• Follow-up
• Miscellaneous
• References

Medical Care

See Further Outpatient Care.

Surgical Care

Surgical care is indicated only in cases in which gastric or esophageal ruptures are suspected.

Consultations

• Cognitive behavior therapy (CBT) has been shown to improve outcomes in patients with bulimia. Referral to a mental health provider with experience in CBT is recommended.
• Dental consultation is appropriate to help avoid progression of enamel erosion due to frequent vomiting.
• Nutrition consultation can be very helpful in guiding patients toward normal eating patterns and healthy food choices.

Diet

• Emphasis in diet instruction is on eating balanced meals, achieving a normal eating pattern, and consuming adequate calories.
• Another goal is to reintroduce feared foods that patients previously binged on in small amounts, allowing patients to enjoy these foods without overeating.

Activity

Patient and family should be counseled to monitor excess activity. Practitioners should stress the importance of playful, pleasurable activities to reduce stress.

MEDICATION

Section 7 of 10  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• Follow-up
• Miscellaneous
• References

The most commonly used medications are antidepressants, typically selective serotonin reuptake inhibitors (SSRIs) or tricyclic antidepressants (TCAs). Evidence in SSRIs is limited to fluoxetine. A Cochrane Database Review showed that patients with bulimia nervosa (BN) who were treated with antidepressants were more likely to interrupt their treatment prematurely due to adverse events.² Patients treated with TCAs dropped out of treatment more frequently than patients treated with placebo. The opposite was found in those treated with fluoxetine, suggesting that it may be a more acceptable treatment. CBT may be a more acceptable first-line treatment than medication, especially in patients concerned about medication side effects.

Drug Category: Selective serotonin reuptake inhibitors (SSRIs)

SSRI medications are thought to help ameliorate depressive symptoms associated with bulimia and to help patients achieve a healthier body image.

SSRIs are greatly preferred over the other classes of antidepressants. Because the adverse effect profile of SSRIs is less prominent, improved compliance is promoted. SSRIs do not have the cardiac arrhythmia risk associated with TCAs. Arrhythmia risk is especially pertinent in overdose, and suicide risk must always be considered when treating a child or adolescent with mood disorder.

Physicians are advised to be aware of the following information and use appropriate caution when considering treatment with SSRIs in the pediatric population.

In December 2003, the UK Medicines and Healthcare Products Regulatory Agency (MHRA) issued an advisory that most SSRIs are not suitable for use by persons younger than 18 years for treatment of "depressive illness." After review, this agency decided that the risks to pediatric patients outweigh the benefits of treatment with SSRIs, except fluoxetine (Prozac), which appears to have a positive risk-benefit ratio in the treatment of depressive illness in patients younger than 18 years.

In October 2003, the US Food and Drug Administration (FDA) issued a public health advisory regarding reports of suicidality in pediatric patients being treated with antidepressant medications for major depressive disorder. This advisory reported suicidality (both ideation and attempts) in clinical trials of various antidepressant drugs in pediatric patients. The FDA has asked that additional studies be performed because suicidality occurred in both treated and untreated patients with major depression and, thus, could not be definitively linked to drug treatment.

However, a study of more than 65,000 children and adults treated for depression from 1992-2002 by the Group Health Cooperative in Seattle found that suicide risk declines, not rises, with the use of antidepressants.³ This is the largest study to date to address this issue.

Currently, evidence does not associate obsessive compulsive disorder (OCD) and other anxiety disorders treated with SSRIs with an increased risk of suicide.

Drug Category: Tricyclic antidepressants

TCAs are thought to help ameliorate depressive symptoms associated with bulimia and to help patients achieve a healthier body image.

Drug Category: Antiemetics

Some studies recommend short-term use of antiemetics at the onset of a patient's treatment. Antiemetics are thought to reduce a patient's stimuli to vomit and help patients through the few weeks it takes for antidepressants to become fully effective.

FOLLOW-UP

Section 8 of 10  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• Follow-up
• Miscellaneous
• References

Further Inpatient Care

• Inpatient care is warranted if patient is suicidal, has abnormal ECG findings or electrolyte levels, is dehydrated, or has had no response to outpatient therapy.
• Inpatient care should include the following:
• • Supervised meals
  • Supervised bathroom privileges
  • No access to bathroom for 2 hours after eating
  • Monitoring of weight and physical activity
  • Assessment of nutritional state
  • Identifications of precipitants to binge and purge
  • Frequent assessment of electrolytes
  • Individual psychotherapy
  • Frequent doctor visits

Further Outpatient Care

• The 2 standard approaches to outpatient care are counseling and medication; these are not mutually exclusive. A combination of these methods has been found to be most effective in patients with bulimia nervosa (BN)
• The most studied form of outpatient care for patients with bulimia is CBT. A specific form of CBT has been created for patients with bulimia and is termed CBT-BN. CBT has been shown to have significantly better results in patients with BN than other forms of psychotherapy.
• • This form of therapy is usually short-term (4-6 mo).
  • CBT focuses on patients' preoccupation with body shape and weight, persistent dieting, and binge eating and purging.
  • Patients are asked to monitor thoughts, feelings, and circumstances surrounding binge-purge episodes. Patients may be asked to keep a food diary and record feelings and urges to binge or purge along with foods the patient consumed during the day. By examining the cues that lead patients to binge, patients can learn to avoid these cues or to redirect their feelings when the cues emerge. These strategies can also help patients challenge their fears of loss of control.
  • Patients are also instructed to cease dieting and begin regular eating. By quitting dieting and removing the feeling of being restricted in what one can eat, patients are less likely to binge on "forbidden foods."
  • Patients are also asked to systematically challenge their assumptions linking weight to self-esteem.
  • Therapy is focused on building trust and developing a treatment alliance.
  • Patients are involved in setting the treatment goals.
  • Some patients benefit from self-help groups. Family involvement in treatment is welcomed and encouraged.
  • The goal of care is to focus on the overall well being of the patient.
• The second approach to treatment is the use of medications.
• • The use of antidepressant medications, such as fluoxetine or a TCA, was initially based on an association between BN and mood disturbance.
  • More than 12 double-blind placebo-controlled trials have shown that antidepressants help patients reduce binge frequency.
• Other staples of outpatient care include nutritional counseling and meal planning.
• Relaxation strategies are helpful for some patients.
• Other forms of therapy with unclear benefit include interpersonal psychotherapy, hypnobehavioral therapy, dialectical behavior therapy, and motivational enhancement therapy. These therapies have not been adequately studied in patients with BN.

In/Out Patient Meds

See Medication.

Deterrence/Prevention

• Prevention efforts have centered on counseling to encourage rational attitudes about weight, moderation of overly high self-expectations, enhancement of self-esteem, and alleviation of stress and stimulating a healthy body image.
• Prevention efforts can be pursued in primary care physicians' offices during health supervision visits.

Complications

• Many complications to bulimia are possible, including the following:
• • CNS - Seizures
  • Cardiac - Cardiac arrhythmias secondary to hypokalemia; can lead to cardiac arrest, cardiac rupture, and cardiomyopathy secondary to ipecac abuse
  • Pulmonary - Pulmonary aspiration of gastric contents, pneumomediastinum
  • GI - Esophageal rupture, esophagitis, delayed gastric emptying, pancreatitis
  • Musculoskeletal - Muscle weakness secondary to ipecac abuse and potassium irregularities, tetany
  • Renal - Impaired renal function
  • Psychiatric - Depression, suicide attempts, substance abuse

Prognosis

• CBT has been shown to benefit patients. Evidence suggests persistent benefit 4 years after treatment; however, treatment benefit greatly depends on accessibility to CBT-trained therapists. Therapists with expertise in CBT may be difficult to find outside of established centers.
• Medication therapy has also been shown to benefit patients; however, only a minority of patients achieve full remission on medication alone. Limited data suggest that a considerable rate of relapse is observed once medications are discontinued.
• Studies have shown that patients who receive treatment (CBT or medication) demonstrate benefit. One study compared treated versus nontreated patients 6 months after initial presentation.⁴ Follow-up studies reported that 28-33% women without treatment were in remission, and follow-up studies of treated women reported 21-75% successful remissions.
• • After one year, follow-up studies reported 28-33% of patients without treatment were in remission, and 5-83% of women in treatment were in remission.
  • Follow-up studies reported a range of 13-69% of patients still in remission without ongoing treatment 2-4 years after initial remission. These data were compared to data pertaining to women who were still in treatment; these women had remission ranges of 46-50%. Over 5 years of follow-up, untreated women had maintained remission rates around 31-60%. This was compared to women who were still in treatment, who had an average remission rate of 54%. Women in treatment outcome studies had higher rates of remission than women in studies who did not receive treatment.
  • Overall, 5-10 years following presentation, approximately 50% of all women with BN fully recover, and 20% still have full BN.
• Given that eating disorders only have been defined and studied as diseases for 20 years, limited long-term data on prognosis exist. The imprecise, inconsistent, and often confusing data on remission rates and percentages available for patient follow-up are evidence that more research and follow-up are needed.
• Overall, despite advances in medical care and therapy, the prognosis for patients with bulimia remains guarded. Even in the best of hands, both medications and therapy fail in 33-50% of patients. The relapse rate remains around 30%, and patient crossover to AN from BN ranges from 0-7%.
• Studies have shown that patients with bulimia who have a previous diagnosis of AN are more likely to have a protracted illness or relapse into AN during follow up compared with patients with bulimia with no history of AN.⁵

Patient Education

• Education of patients and families involves teaching the seriousness and consequences of bulimic behavior.
• Information about complications and physiologic changes that can occur as a result of bulimia is important to convey to patients and families.
• Information on proper nutrition and metabolic balance is also helpful.
• For excellent patient education resources, visit eMedicine's Eating Disorders Center. Also, see eMedicine's patient education article Bulimia.

MISCELLANEOUS

Section 9 of 10  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• Follow-up
• Miscellaneous
• References

Medical/Legal Pitfalls

• Because many patients with bulimia nervosa (BN) have concomitant psychiatric disorders, screening patients for suicidal intentions is worthwhile.

Special Concerns

• One research group in the United Kingdom has designed a screening tool to be used for patients in whom bulimia is suspected.⁶ This questionnaire, called the SCOFF (sick, control, one, fat, food) questionnaire, is designed to be similar to the CAGE ([need to] cut down [on drinking], annoyance, guilt [about drinking], [need for] eye-opener) questionnaire used in alcoholism screening. These 5 questions are designed to address core features of AN and BN. The false-positive result rate is approximately 12.5%. The test has very high sensitivity. One point per yes answer is added, and a score greater than 2 indicates a likely case of AN or BN. The questions are as follows:
• • Do you make yourself sick because you feel uncomfortably full? 
  • Do you worry you have lost control over how much you eat?
  • Have you recently lost more than one stone in a 3-month period? (A stone is a United Kingdom measurement of weight and equals approximately 14 lb.)
  • Do you believe yourself to be fat when others say you are too thin?
  • Would you say that food dominates your life?

REFERENCES

Section 10 of 10

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• Follow-up
• Miscellaneous
• References

1. Fedorowicz VJ, Falissard B, Foulon C, Dardennes R, Divac SM, Guelfi JD, et al. Factors associated with suicidal behaviors in a large French sample of inpatients with eating disorders. Int J Eat Disord. Nov 2007;40(7):589-95. [Medline]. [Full Text].
2. Bacaltchuk J, Hay P. Antidepressants versus placebo for people with bulimia nervosa.[update of Cochrane Database Syst Rev. 2001]. Cochrane Database of Systematic Reviews. 2003;4:[Full Text].
3. Simon GE, Savarino J, Operskalski B, Wang PS. Suicide risk during antidepressant treatment. Am J Psychiatry. Jan 2006;163(1):41-7. [Medline]. [Full Text].
4. Keel PK, Mitchell JE. Outcome in bulimia nervosa. Am J Psychiatry. Mar 1997;154(3):313-21. [Medline]. [Full Text].
5. Eddy KT, Dorer DJ, Franko DL, Tahilani K, Thompson-Brenner H, Herzog DB. Should bulimia nervosa be subtyped by history of anorexia nervosa? A longitudinal validation. Int J Eat Disord. Nov 2007;40 Suppl:S67-71. [Medline]. [Full Text].
6. Morgan JF, Reid F, Lacey JH. The SCOFF questionnaire: assessment of a new screening tool for eating disorders. BMJ. Dec 4 1999;319(7223):1467-8. [Medline]. [Full Text].
7. Behrman RE, Kliegman R, Jenson HB. Nelson's Textbook of Pediatrics. 16^th ed. Philadelphia, Pa: WB Saunders and Co; 2000.
8. Carney CP, Andersen AE. Eating disorders. Guide to medical evaluation and complications. Psychiatr Clin North Am. Dec 1996;19(4):657-79. [Medline].
9. Cotran RS, Kumar V, Collins T. Robbins Pathologic Basis of Disease. 6^th ed. Philadelphia, Pa: WB Saunders and Co; 1999.
10. Dambro MR, Griffith JA, Griffith HG. Griffith's Five Minute Clinical Consult. 5^th ed. Baltimore, Md: Williams & Wilkins; 1999.
11. Ellenhorn MJ. Ellenhorn's Medical Toxicology. 2^nd ed. Baltimore, Md: Williams & Wilkins; 1997.
12. Halmi KA. Models to conceptualize risk factors for bulimia nervosa. Arch Gen Psychiatry. Jun 1997;54(6):507-8. [Medline].
13. Haly PJ, Bacltchuck J. Bulimia Nervosa. American Family Physician. June 2007;75:1699-702.
14. Hartman BK, Faris PL, Kim SW, Raymond NC, Goodale RL, Meller WH, et al. Treatment of bulimia nervosa with ondansetron. Arch Gen Psychiatry. Oct 1997;54(10):969-70. [Medline].
15. Hay PJ, Bacaltchuk J, Stefano S. Psychotherapy for bulimia nervosa and binging. Cochrane Database Syst Rev. 2004;(3):CD000562. [Medline]. [Full Text].
16. Lilly RZ. Bulimia nervosa. BMJ. Aug 16 2003;327(7411):380-1. [Medline]. [Full Text].
17. Muise AM, Stein DG, Arbess G. Eating disorders in adolescent boys: a review of the adolescent and young adult literature. J Adolesc Health. Dec 2003;33(6):427-35. [Medline].
18. Nicholls D, Viner R. Eating disorders and weight problems. BMJ. Apr 23 2005;330(7497):950-3. [Medline]. [Full Text].
19. Siberry GK, Iannone R. The Harriet Lane Handbook: A Manual for Pediatric House Officers. 15^th ed. Mosby-Year Book; 2000.
20. Spindler A, Milos G. Links between eating disorder symptom severity and psychiatric comorbidity. Eat Behav. Aug 2007;8(3):364-73. [Medline].
21. Taketomo CK, Hodding JH, Kraus DM. Pediatric Dosage Handbook. 6^th ed. 1999.
22. Walsh BT, Devlin MJ. Eating disorders: progress and problems. Science. May 29 1998;280(5368):1387-90. [Medline].

Article Last Updated: Feb 25, 2008