Paul E. Keck Jr, MD

Because no one treatment fits all patients with binge eating disorder, their management usually requires an individualized program of:

• behavioral weight control
• psychotherapy
• and sometimes medications.

In our weight management clinic, we consider medication options based on patient preference and whether BED is uncomplicated (Figure 1) or coexists with a mood disorder (Figure 2).

This article presents the evidence on which we base our comprehensive approach. General psychiatrists with knowledge of BED can treat patients with this eating disorder, although complicated cases may require referral for specialized treatment.

Figure 1 Medication options for uncomplicated BED

Clinical characteristics

Psychiatric comorbidity. BED often occurs in patients with mood, anxiety, substance-abuse, impulsecontrol, and personality disorders.^4,6,10-12 Mood disorder—particularly depression—appears to be the most common comorbidity. BED can occur with bipolar disorder¹²—a comorbidity that in our experience is underrecognized both clinically and in the literature.

Patients with BED and bipolar disorder show increased impulsivity and mood lability. As bipolar II disorder and other “soft-spectrum” forms are more common than bipolar I disorder, BED is also more likely to occur with hypomania than mania.

Overweight. Not surprisingly, BED is associated with overweight and obesity.^5,8,9,11 Not all patients with BED are overweight or obese, but most who participate in clinical trials of BED treatments are at least overweight. BED has been reported in up to:

• 30% of participants in weight-loss programs⁷
• 70% of participants in groups such as Overeaters Anonymous
• 50% of patients who seek bariatric surgery.⁵

In our experience, patients are often more distressed by their weight than by their binge eating, depression, or anxiety. Indeed, overweight and obesity are the usual reasons patients with BED present for treatment at our center.

Diagnosis. BED’s validity as a clinical diagnosis has been controversial since the disorder was first included in DSM-IV (Table 1).³ Debate continues about some definitions in the DSM criteria, including what amount of food is “definitely larger” than most people would eat and what is “loss of control over eating.”

Nevertheless, screening for BED is relatively easy. Clinicians may use the eating disorder section of the Structured Clinical Interview for DSM-IV or the Eating Disorders Examination. Alternatively, simply ask patients if they have episodes of uncontrollable overeating, during which they eat unusually large amounts of food and their eating feels out of control.

Course. BED begins in adolescence or adulthood. Disease course is variable, with periods of remission, recurrence, and chronicity.^6,7,10 Interestingly, one prospective study showed that even if the binge eating resolves, persons may still develop obesity.¹³

Prevalence. BED affects 1.5% to 3% of the U.S. population. It is more common in women than men, equally prevalent in whites and blacks, and more prevalent than anorexia nervosa and bulimia nervosa combined.^11,14 Subthreshold BED—such as obesity with infrequent or nondistressing binge eating—appears to be much more common,¹⁰ although no data are available.

Theories of binge eating

BED’s cause is unknown, but biological, familial, and psychosocial factors have been implicated.

Biological factors. The neurotransmitters serotonin (5-HT) and dopamine—as well as various peptides—have been shown to help regulate feeding behavior.¹⁰

Table 1

Diagnostic criteria for binge-eating disorder*

Serotonin. Reduced 5-HT transporter binding has been shown in obese women with BED.¹⁵ Their 5-HT binding improved and binge eating subsided with group psychotherapy and fluoxetine, although the women continued to gain weight.

Figure 2 Medication options for BED with obesity and a mood disorder*

Genetic factors. In severely obese patients (body mass index 44±2), those with a DSM-IV diagnosis of BED exhibited mutations of the melanocortin 4 receptor gene, which affects the anorectic properties of alpha melanocyte-stimulating hormone.¹⁷

Familial factors associated with BED include parental depression and obesity.¹⁸

Psychosocial correlates include physical and sexual abuse, bullying by peers, and discrimination because of being overweight.¹⁹

Treatment recommendations

Few systematic studies have examined BED treatment. Emerging research suggests that behavioral weight-loss treatment, specialized psychotherapies, and medications may be effective in some patients with BED.^4,6,8

Behavioral weight-loss treatment’s main goal is to manage the patient’s weight with a lower-calorie, healthy diet and to increase exercise.^20,21

Over the short term (<1 year), behavioral weight-loss treatment produces similar weight loss in obese patients with or without BED; long-term results in both groups, however, have not been satisfactory.^20,21 No studies have examined the efficacy of specialized diets (such as low-carbohydrate regimens) in patients with BED.

Specialized psychotherapy’s goal is to modify bingeeating behavior with behavioral self-management strategies, reducing interpersonal dysfunction and stress, and/or managing affective dysregulation.

Cognitive-behavioral therapy (CBT) and interpersonal therapy (IPT) have been effective in reducing binge eating, both acutely and for up to 12 months^4,20-24 but less effective in achieving and maintaining weight loss. Patients who achieve remission in binge eating after undergoing CBT or IPI often experience modest but stable weight loss.^20-22 For example, in a comparison study of CBT and IPT:

• After 20 weekly sessions, patients whose binge eating was in remission lost weight (mean body mass index [BMI] −0.5 ± 1.5 kg/m²), whereas those who continued to binge gained weight (mean BMI +0.4 ±2.0 kg/m²).
• At 12 months’ follow-up, patients still in remission continued to lose weight (mean BMI −1.0 ± 3.0 kg/m²), whereas those no longer in remission gained weight (mean BMI +0.7 ±2.9 kg/m²[P = 0.01]).²²

Self-help and dialectical behavioral therapy (DBT) may also help reduce binge eating in BED. As with CBT and IPT, they are less effective in weight loss. In the only controlled study of DBT,²⁴ patients achieved an average 2.5-lb weight loss after 20 weeks of DBT, compared with an average 0.6-lb weight gain in the control group. This difference was not significant, and the report did not include data on weight loss maintenance.

In summary, CBT may be more effective than behavioral weight loss treatment for reducing binge eating, but behavioral weight loss is more effective for weight loss.

Medications for BED

Medications that have been tried for BED include antidepressants, appetite suppressants, and anticonvulsants.^25,26 Antidepressants are used to treat BED because:

• BED is often associated with depressive symptoms and disorders.
• BED is related to bulimia nervosa, and placebo-controlled trials have shown that the binge eating of bulimia nervosa responds to several classes of antidepressants. The selective serotonin reuptake inhibitor (SSRI) fluoxetine is the only medication indicated for treating any eating disorder (bulimia nervosa).
• Bupropion and venlafaxine—a serotonin-norepinephrine reuptake inhibitor (SNRI)—have weight-loss properties.

SSRIs are the most extensively studied antidepressants for treating BED. SSRIs have weightloss properties, but only short term.^25-26 Citalopram, fluoxetine, fluvoxamine, and sertraline have reduced binge eating and body weight more effectively than placebo during 6 to 9 weeks of treatment (Table 2).^25-26 However, one controlled study²³ showed that fluoxetine was not significantly more effective than placebo in reducing binge frequency or body weight after 16 weeks.

TCAs. Studies of tricyclic antidepressants (TCAs) for BED are sparse, and results have been mixed. In one trial, imipramine was similar to placebo in reducing binge frequency and body weight. In a placebo-controlled study of patients with nonpurging bulimia nervosa, desipramine reduced binge eating but had no effect on body weight.^25,26

Table 2

Drug therapies shown to be effective for BED*

Venlafaxine. In a retrospective review of 35 consecutive obese women with BED, venlafaxine, mean 222 mg/d for 28 to 300 days (median 120 days), reduced binge eating, body weight, and depressive symptoms.²⁷

Bupropion has been more effective than placebo for treating:

• uncomplicated obesity (short- and long-term)
• obesity associated with depressive symptoms
• bulimia nervosa (although bupropion is contraindicated in these patients because of seizure risk).^26,28,29

No controlled trials have studied bupropion for BED. When using dosages effective in depressive disorders, we find bupropion helpful in reducing binge eating, body weight, and depressive symptoms in BED patients.

Appetite suppressants decrease appetite and weight, may increase satiety, and may reduce depressive symptoms.

Sibutramine—a serotonin, norepinephrine, and dopamine reuptake inhibitor indicated for managing obesity—has been reported effective in BED in a 12-week, randomized, double-blind, placebo-controlled trial. A 15-mg/d dosage reduced binge frequency, body weight, and depressive symptoms more effectively than placebo in 60 obese patients with BED.³⁰ Most-frequent adverse effects (dry mouth and constipation) were mild and benign, and no significant complications were observed.

Sibutramine’s mechanism of action in BED is unknown. However, it suppressed food intake during binge-eating episodes in patients with BED in a randomized, controlled, cross-over laboratory study.³¹

Orlistat. We know of no published controlled studies of the lipase inhibitor orlistat in treating BED. In our experience, some patients do well with this agent, though we have observed infrequent purging episodes with it in patients with BED.

With orlistat, 120 mg tid, our BED patients have experienced weight loss comparable to that seen in uncomplicated obesity at similar dosages. Orlistat seems most effective for:

• patients whose binge eating is in remission
• those who responded to behavioral weightloss treatment, a psychological treatment, or another medication.

Anticonvulsants such as topiramate and zonisamide have been shown effective in treating obesity^32,33 and are sometimes used to treat BED. Obese BED patients with mood disorders often do best with psychotherapy plus medication

Topiramate at dosages of 50 to 600 mg/d (median 212 mg/d) reduced binge-eating frequency, obsessive-compulsive features of binge eating, and body weight more effectively than placebo in a 14-week study of 61 obese patients with BED. These effects were maintained across 48 weeks in an open-label extension trial.³⁴

Zonisamide, mean 513 mg/d, produced similar results during a prospective, open-label, 12-week trial in 15 patients with BED.³⁵ A controlled trial to replicate these findings is ongoing.

BED may respond to anticonvulsant therapy for several reasons:

• Some anticonvulsants are effective in treating bipolar disorder, which may occur with BED.¹²
• Some anticonvulsants have shown benefit in conditions associated with pathologic impulsivity, such as substance abuse, impulse-control, and cluster B disorders.¹⁰

Growing evidence shows that bulimia nervosa and BED may be associated with pathologic impulsivity.

Combination therapies are generally more effective than monotherapies in patients with mood disorders, uncomplicated obesity, and possibly bulimia nervosa. Even so, few trials have systematically studied combination therapy in managing patients with BED.

Two studies compared psychotherapy and antidepressants alone and in combination in treating BED.^21,23 Both showed that CBT alone was more effective in decreasing binge frequency than desipramine alone,²¹ fluoxetine alone,²³ and the combination of CBT and medication. On the other hand, patients who took desipramine either alone or in combination experienced a greater degree of weight loss than those who did not take desipramine.²¹

In another combination therapy, exercise has been shown to be an effective adjunct to CBT in maintenance treatment of obese women with BED.³⁶

No studies have compared behavioral weight management or a specialized psychotherapy in combination with an antiobesity drug or a weight-loss anticonvulsant in treating BED.

Treatment recommendations

In our experience, BED patients—particularly those with obesity and psychopathology—often do best with some combination of psychological treatment and medication:

• The psychological component may be behavioral weight-loss treatment, a specialized psychotherapy such as CBT or IPT, or some combination of behavioral weight-loss treatment and specialized psychotherapy.
• The medication component may consist of an antidepressant, anticonvulsant, antiobesity drug, or multiple drugs (such as an SSRI or sibutramine with topiramate for BED with major depression, or topiramate with lithium for BED with bipolar disorder).

Although combination therapies may be optimal for some patients, this approach remains unproven in controlled trials.

Patient preference. In addition to comorbidities, patient preference is an important consideration when choosing BED treatments. We determine our patients’ preferences by educating them as much as possible about their options. We explain the benefits and weaknesses of all treatments and encourage them to participate in forming their individualized treatment plans.

Patients sometimes have strong treatment preferences. Some prefer psychological treatments, whereas others prefer medications. Working with patient preferences enhances treatment adherence. For example, patients who fail a preferred treatment are often more willing to adhere to another treatment modality about which they initially were skeptical.

Related resources

• Bray GA, Bouchard C (eds). Handbook of obesity: clinical applications (2nd ed). New York, NY: Marcel Dekker, 2004.
• Cooper Z, Zairburn CG, Hawker DM. Cognitive behavioral treatment of obesity. A clinician’s guide. New York: Guilford Press, 2003.
• Carter WP, Hudson JI, Lalonde JK, et al. Pharmacologic treatment of binge eating disorder. Int J Eat Disord 2003;34(suppl):S74-88.

Drug brand names

• Bupropion • Wellbutrin
• Citalopram • Celexa
• Desipramine • Norpramin
• Fluoxetine • Prozac
• Fluvoxamine • Luvox
• Imipramine • Tofranil
• Lamotrigine • Lamictal
• Lithium • Eskalith, others
• Orlistat • Xenical
• Sertraline • Zoloft
• Sibutramine • Meridia
• Topiramate • Topamax
• Venlafaxine • Effexor
• Zonisamide • Zonegran

Disclosure

Dr. Kotwal receives grant support from Elan Corporation and is a speaker for Ortho-McNeil Pharmaceutical and Pfizer Inc.

Dr. Kaneria and Ms. Guerdjikova report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Dr. McElroy is a consultant to Abbott Laboratories, Bristol-Myers Squibb Co., Elan Corporation, GlaxoSmithKline, Janssen Pharmaceutica, Eli Lilly and Co., and Ortho-McNeil Pharmaceutical. She receives grant/research support from Elan Pharmaceuticals, Forest Pharmaceuticals, Merck & Co., Ortho-McNeil Pharmaceutical, and Sanofi-Synthelabo and is a speaker for Eli Lilly and Co. and Ortho-McNeil Pharmaceutical.

Because no one treatment fits all patients with binge eating disorder, their management usually requires an individualized program of:

• behavioral weight control
• psychotherapy
• and sometimes medications.

In our weight management clinic, we consider medication options based on patient preference and whether BED is uncomplicated (Figure 1) or coexists with a mood disorder (Figure 2).

This article presents the evidence on which we base our comprehensive approach. General psychiatrists with knowledge of BED can treat patients with this eating disorder, although complicated cases may require referral for specialized treatment.

Figure 1 Medication options for uncomplicated BED

Clinical characteristics

Psychiatric comorbidity. BED often occurs in patients with mood, anxiety, substance-abuse, impulsecontrol, and personality disorders.^4,6,10-12 Mood disorder—particularly depression—appears to be the most common comorbidity. BED can occur with bipolar disorder¹²—a comorbidity that in our experience is underrecognized both clinically and in the literature.

Patients with BED and bipolar disorder show increased impulsivity and mood lability. As bipolar II disorder and other “soft-spectrum” forms are more common than bipolar I disorder, BED is also more likely to occur with hypomania than mania.

Overweight. Not surprisingly, BED is associated with overweight and obesity.^5,8,9,11 Not all patients with BED are overweight or obese, but most who participate in clinical trials of BED treatments are at least overweight. BED has been reported in up to:

• 30% of participants in weight-loss programs⁷
• 70% of participants in groups such as Overeaters Anonymous
• 50% of patients who seek bariatric surgery.⁵

In our experience, patients are often more distressed by their weight than by their binge eating, depression, or anxiety. Indeed, overweight and obesity are the usual reasons patients with BED present for treatment at our center.

Diagnosis. BED’s validity as a clinical diagnosis has been controversial since the disorder was first included in DSM-IV (Table 1).³ Debate continues about some definitions in the DSM criteria, including what amount of food is “definitely larger” than most people would eat and what is “loss of control over eating.”

Nevertheless, screening for BED is relatively easy. Clinicians may use the eating disorder section of the Structured Clinical Interview for DSM-IV or the Eating Disorders Examination. Alternatively, simply ask patients if they have episodes of uncontrollable overeating, during which they eat unusually large amounts of food and their eating feels out of control.

Course. BED begins in adolescence or adulthood. Disease course is variable, with periods of remission, recurrence, and chronicity.^6,7,10 Interestingly, one prospective study showed that even if the binge eating resolves, persons may still develop obesity.¹³

Prevalence. BED affects 1.5% to 3% of the U.S. population. It is more common in women than men, equally prevalent in whites and blacks, and more prevalent than anorexia nervosa and bulimia nervosa combined.^11,14 Subthreshold BED—such as obesity with infrequent or nondistressing binge eating—appears to be much more common,¹⁰ although no data are available.

Theories of binge eating

BED’s cause is unknown, but biological, familial, and psychosocial factors have been implicated.

Biological factors. The neurotransmitters serotonin (5-HT) and dopamine—as well as various peptides—have been shown to help regulate feeding behavior.¹⁰

Table 1

Diagnostic criteria for binge-eating disorder*

Serotonin. Reduced 5-HT transporter binding has been shown in obese women with BED.¹⁵ Their 5-HT binding improved and binge eating subsided with group psychotherapy and fluoxetine, although the women continued to gain weight.

Figure 2 Medication options for BED with obesity and a mood disorder*

Genetic factors. In severely obese patients (body mass index 44±2), those with a DSM-IV diagnosis of BED exhibited mutations of the melanocortin 4 receptor gene, which affects the anorectic properties of alpha melanocyte-stimulating hormone.¹⁷

Familial factors associated with BED include parental depression and obesity.¹⁸

Psychosocial correlates include physical and sexual abuse, bullying by peers, and discrimination because of being overweight.¹⁹

Treatment recommendations

Few systematic studies have examined BED treatment. Emerging research suggests that behavioral weight-loss treatment, specialized psychotherapies, and medications may be effective in some patients with BED.^4,6,8

Behavioral weight-loss treatment’s main goal is to manage the patient’s weight with a lower-calorie, healthy diet and to increase exercise.^20,21

Over the short term (<1 year), behavioral weight-loss treatment produces similar weight loss in obese patients with or without BED; long-term results in both groups, however, have not been satisfactory.^20,21 No studies have examined the efficacy of specialized diets (such as low-carbohydrate regimens) in patients with BED.

Specialized psychotherapy’s goal is to modify bingeeating behavior with behavioral self-management strategies, reducing interpersonal dysfunction and stress, and/or managing affective dysregulation.

Cognitive-behavioral therapy (CBT) and interpersonal therapy (IPT) have been effective in reducing binge eating, both acutely and for up to 12 months^4,20-24 but less effective in achieving and maintaining weight loss. Patients who achieve remission in binge eating after undergoing CBT or IPI often experience modest but stable weight loss.^20-22 For example, in a comparison study of CBT and IPT:

• After 20 weekly sessions, patients whose binge eating was in remission lost weight (mean body mass index [BMI] −0.5 ± 1.5 kg/m²), whereas those who continued to binge gained weight (mean BMI +0.4 ±2.0 kg/m²).
• At 12 months’ follow-up, patients still in remission continued to lose weight (mean BMI −1.0 ± 3.0 kg/m²), whereas those no longer in remission gained weight (mean BMI +0.7 ±2.9 kg/m²[P = 0.01]).²²

Self-help and dialectical behavioral therapy (DBT) may also help reduce binge eating in BED. As with CBT and IPT, they are less effective in weight loss. In the only controlled study of DBT,²⁴ patients achieved an average 2.5-lb weight loss after 20 weeks of DBT, compared with an average 0.6-lb weight gain in the control group. This difference was not significant, and the report did not include data on weight loss maintenance.

In summary, CBT may be more effective than behavioral weight loss treatment for reducing binge eating, but behavioral weight loss is more effective for weight loss.

Medications for BED

Medications that have been tried for BED include antidepressants, appetite suppressants, and anticonvulsants.^25,26 Antidepressants are used to treat BED because:

• BED is often associated with depressive symptoms and disorders.
• BED is related to bulimia nervosa, and placebo-controlled trials have shown that the binge eating of bulimia nervosa responds to several classes of antidepressants. The selective serotonin reuptake inhibitor (SSRI) fluoxetine is the only medication indicated for treating any eating disorder (bulimia nervosa).
• Bupropion and venlafaxine—a serotonin-norepinephrine reuptake inhibitor (SNRI)—have weight-loss properties.

SSRIs are the most extensively studied antidepressants for treating BED. SSRIs have weightloss properties, but only short term.^25-26 Citalopram, fluoxetine, fluvoxamine, and sertraline have reduced binge eating and body weight more effectively than placebo during 6 to 9 weeks of treatment (Table 2).^25-26 However, one controlled study²³ showed that fluoxetine was not significantly more effective than placebo in reducing binge frequency or body weight after 16 weeks.

TCAs. Studies of tricyclic antidepressants (TCAs) for BED are sparse, and results have been mixed. In one trial, imipramine was similar to placebo in reducing binge frequency and body weight. In a placebo-controlled study of patients with nonpurging bulimia nervosa, desipramine reduced binge eating but had no effect on body weight.^25,26

Table 2

Drug therapies shown to be effective for BED*

Venlafaxine. In a retrospective review of 35 consecutive obese women with BED, venlafaxine, mean 222 mg/d for 28 to 300 days (median 120 days), reduced binge eating, body weight, and depressive symptoms.²⁷

Bupropion has been more effective than placebo for treating:

• uncomplicated obesity (short- and long-term)
• obesity associated with depressive symptoms
• bulimia nervosa (although bupropion is contraindicated in these patients because of seizure risk).^26,28,29

No controlled trials have studied bupropion for BED. When using dosages effective in depressive disorders, we find bupropion helpful in reducing binge eating, body weight, and depressive symptoms in BED patients.

Appetite suppressants decrease appetite and weight, may increase satiety, and may reduce depressive symptoms.

Sibutramine—a serotonin, norepinephrine, and dopamine reuptake inhibitor indicated for managing obesity—has been reported effective in BED in a 12-week, randomized, double-blind, placebo-controlled trial. A 15-mg/d dosage reduced binge frequency, body weight, and depressive symptoms more effectively than placebo in 60 obese patients with BED.³⁰ Most-frequent adverse effects (dry mouth and constipation) were mild and benign, and no significant complications were observed.

Sibutramine’s mechanism of action in BED is unknown. However, it suppressed food intake during binge-eating episodes in patients with BED in a randomized, controlled, cross-over laboratory study.³¹

Orlistat. We know of no published controlled studies of the lipase inhibitor orlistat in treating BED. In our experience, some patients do well with this agent, though we have observed infrequent purging episodes with it in patients with BED.

With orlistat, 120 mg tid, our BED patients have experienced weight loss comparable to that seen in uncomplicated obesity at similar dosages. Orlistat seems most effective for:

• patients whose binge eating is in remission
• those who responded to behavioral weightloss treatment, a psychological treatment, or another medication.

Anticonvulsants such as topiramate and zonisamide have been shown effective in treating obesity^32,33 and are sometimes used to treat BED. Obese BED patients with mood disorders often do best with psychotherapy plus medication

Topiramate at dosages of 50 to 600 mg/d (median 212 mg/d) reduced binge-eating frequency, obsessive-compulsive features of binge eating, and body weight more effectively than placebo in a 14-week study of 61 obese patients with BED. These effects were maintained across 48 weeks in an open-label extension trial.³⁴

Zonisamide, mean 513 mg/d, produced similar results during a prospective, open-label, 12-week trial in 15 patients with BED.³⁵ A controlled trial to replicate these findings is ongoing.

BED may respond to anticonvulsant therapy for several reasons:

• Some anticonvulsants are effective in treating bipolar disorder, which may occur with BED.¹²
• Some anticonvulsants have shown benefit in conditions associated with pathologic impulsivity, such as substance abuse, impulse-control, and cluster B disorders.¹⁰

Growing evidence shows that bulimia nervosa and BED may be associated with pathologic impulsivity.

Combination therapies are generally more effective than monotherapies in patients with mood disorders, uncomplicated obesity, and possibly bulimia nervosa. Even so, few trials have systematically studied combination therapy in managing patients with BED.

Two studies compared psychotherapy and antidepressants alone and in combination in treating BED.^21,23 Both showed that CBT alone was more effective in decreasing binge frequency than desipramine alone,²¹ fluoxetine alone,²³ and the combination of CBT and medication. On the other hand, patients who took desipramine either alone or in combination experienced a greater degree of weight loss than those who did not take desipramine.²¹

In another combination therapy, exercise has been shown to be an effective adjunct to CBT in maintenance treatment of obese women with BED.³⁶

No studies have compared behavioral weight management or a specialized psychotherapy in combination with an antiobesity drug or a weight-loss anticonvulsant in treating BED.

Treatment recommendations

In our experience, BED patients—particularly those with obesity and psychopathology—often do best with some combination of psychological treatment and medication:

• The psychological component may be behavioral weight-loss treatment, a specialized psychotherapy such as CBT or IPT, or some combination of behavioral weight-loss treatment and specialized psychotherapy.
• The medication component may consist of an antidepressant, anticonvulsant, antiobesity drug, or multiple drugs (such as an SSRI or sibutramine with topiramate for BED with major depression, or topiramate with lithium for BED with bipolar disorder).

Although combination therapies may be optimal for some patients, this approach remains unproven in controlled trials.

Patient preference. In addition to comorbidities, patient preference is an important consideration when choosing BED treatments. We determine our patients’ preferences by educating them as much as possible about their options. We explain the benefits and weaknesses of all treatments and encourage them to participate in forming their individualized treatment plans.

Patients sometimes have strong treatment preferences. Some prefer psychological treatments, whereas others prefer medications. Working with patient preferences enhances treatment adherence. For example, patients who fail a preferred treatment are often more willing to adhere to another treatment modality about which they initially were skeptical.

Related resources

• Bray GA, Bouchard C (eds). Handbook of obesity: clinical applications (2nd ed). New York, NY: Marcel Dekker, 2004.
• Cooper Z, Zairburn CG, Hawker DM. Cognitive behavioral treatment of obesity. A clinician’s guide. New York: Guilford Press, 2003.
• Carter WP, Hudson JI, Lalonde JK, et al. Pharmacologic treatment of binge eating disorder. Int J Eat Disord 2003;34(suppl):S74-88.

Drug brand names

• Bupropion • Wellbutrin
• Citalopram • Celexa
• Desipramine • Norpramin
• Fluoxetine • Prozac
• Fluvoxamine • Luvox
• Imipramine • Tofranil
• Lamotrigine • Lamictal
• Lithium • Eskalith, others
• Orlistat • Xenical
• Sertraline • Zoloft
• Sibutramine • Meridia
• Topiramate • Topamax
• Venlafaxine • Effexor
• Zonisamide • Zonegran

Disclosure

Dr. Kotwal receives grant support from Elan Corporation and is a speaker for Ortho-McNeil Pharmaceutical and Pfizer Inc.

Dr. Kaneria and Ms. Guerdjikova report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Dr. McElroy is a consultant to Abbott Laboratories, Bristol-Myers Squibb Co., Elan Corporation, GlaxoSmithKline, Janssen Pharmaceutica, Eli Lilly and Co., and Ortho-McNeil Pharmaceutical. She receives grant/research support from Elan Pharmaceuticals, Forest Pharmaceuticals, Merck & Co., Ortho-McNeil Pharmaceutical, and Sanofi-Synthelabo and is a speaker for Eli Lilly and Co. and Ortho-McNeil Pharmaceutical.

Because no one treatment fits all patients with binge eating disorder, their management usually requires an individualized program of:

• behavioral weight control
• psychotherapy
• and sometimes medications.

In our weight management clinic, we consider medication options based on patient preference and whether BED is uncomplicated (Figure 1) or coexists with a mood disorder (Figure 2).

This article presents the evidence on which we base our comprehensive approach. General psychiatrists with knowledge of BED can treat patients with this eating disorder, although complicated cases may require referral for specialized treatment.

Figure 1 Medication options for uncomplicated BED

Clinical characteristics

Psychiatric comorbidity. BED often occurs in patients with mood, anxiety, substance-abuse, impulsecontrol, and personality disorders.^4,6,10-12 Mood disorder—particularly depression—appears to be the most common comorbidity. BED can occur with bipolar disorder¹²—a comorbidity that in our experience is underrecognized both clinically and in the literature.

Patients with BED and bipolar disorder show increased impulsivity and mood lability. As bipolar II disorder and other “soft-spectrum” forms are more common than bipolar I disorder, BED is also more likely to occur with hypomania than mania.

Overweight. Not surprisingly, BED is associated with overweight and obesity.^5,8,9,11 Not all patients with BED are overweight or obese, but most who participate in clinical trials of BED treatments are at least overweight. BED has been reported in up to:

• 30% of participants in weight-loss programs⁷
• 70% of participants in groups such as Overeaters Anonymous
• 50% of patients who seek bariatric surgery.⁵

In our experience, patients are often more distressed by their weight than by their binge eating, depression, or anxiety. Indeed, overweight and obesity are the usual reasons patients with BED present for treatment at our center.

Diagnosis. BED’s validity as a clinical diagnosis has been controversial since the disorder was first included in DSM-IV (Table 1).³ Debate continues about some definitions in the DSM criteria, including what amount of food is “definitely larger” than most people would eat and what is “loss of control over eating.”

Nevertheless, screening for BED is relatively easy. Clinicians may use the eating disorder section of the Structured Clinical Interview for DSM-IV or the Eating Disorders Examination. Alternatively, simply ask patients if they have episodes of uncontrollable overeating, during which they eat unusually large amounts of food and their eating feels out of control.

Course. BED begins in adolescence or adulthood. Disease course is variable, with periods of remission, recurrence, and chronicity.^6,7,10 Interestingly, one prospective study showed that even if the binge eating resolves, persons may still develop obesity.¹³

Prevalence. BED affects 1.5% to 3% of the U.S. population. It is more common in women than men, equally prevalent in whites and blacks, and more prevalent than anorexia nervosa and bulimia nervosa combined.^11,14 Subthreshold BED—such as obesity with infrequent or nondistressing binge eating—appears to be much more common,¹⁰ although no data are available.

Theories of binge eating

BED’s cause is unknown, but biological, familial, and psychosocial factors have been implicated.

Biological factors. The neurotransmitters serotonin (5-HT) and dopamine—as well as various peptides—have been shown to help regulate feeding behavior.¹⁰

Table 1

Diagnostic criteria for binge-eating disorder*

Serotonin. Reduced 5-HT transporter binding has been shown in obese women with BED.¹⁵ Their 5-HT binding improved and binge eating subsided with group psychotherapy and fluoxetine, although the women continued to gain weight.

Figure 2 Medication options for BED with obesity and a mood disorder*

Genetic factors. In severely obese patients (body mass index 44±2), those with a DSM-IV diagnosis of BED exhibited mutations of the melanocortin 4 receptor gene, which affects the anorectic properties of alpha melanocyte-stimulating hormone.¹⁷

Familial factors associated with BED include parental depression and obesity.¹⁸

Psychosocial correlates include physical and sexual abuse, bullying by peers, and discrimination because of being overweight.¹⁹

Treatment recommendations

Few systematic studies have examined BED treatment. Emerging research suggests that behavioral weight-loss treatment, specialized psychotherapies, and medications may be effective in some patients with BED.^4,6,8

Behavioral weight-loss treatment’s main goal is to manage the patient’s weight with a lower-calorie, healthy diet and to increase exercise.^20,21

Over the short term (<1 year), behavioral weight-loss treatment produces similar weight loss in obese patients with or without BED; long-term results in both groups, however, have not been satisfactory.^20,21 No studies have examined the efficacy of specialized diets (such as low-carbohydrate regimens) in patients with BED.

Specialized psychotherapy’s goal is to modify bingeeating behavior with behavioral self-management strategies, reducing interpersonal dysfunction and stress, and/or managing affective dysregulation.

Cognitive-behavioral therapy (CBT) and interpersonal therapy (IPT) have been effective in reducing binge eating, both acutely and for up to 12 months^4,20-24 but less effective in achieving and maintaining weight loss. Patients who achieve remission in binge eating after undergoing CBT or IPI often experience modest but stable weight loss.^20-22 For example, in a comparison study of CBT and IPT:

• After 20 weekly sessions, patients whose binge eating was in remission lost weight (mean body mass index [BMI] −0.5 ± 1.5 kg/m²), whereas those who continued to binge gained weight (mean BMI +0.4 ±2.0 kg/m²).
• At 12 months’ follow-up, patients still in remission continued to lose weight (mean BMI −1.0 ± 3.0 kg/m²), whereas those no longer in remission gained weight (mean BMI +0.7 ±2.9 kg/m²[P = 0.01]).²²

Self-help and dialectical behavioral therapy (DBT) may also help reduce binge eating in BED. As with CBT and IPT, they are less effective in weight loss. In the only controlled study of DBT,²⁴ patients achieved an average 2.5-lb weight loss after 20 weeks of DBT, compared with an average 0.6-lb weight gain in the control group. This difference was not significant, and the report did not include data on weight loss maintenance.

In summary, CBT may be more effective than behavioral weight loss treatment for reducing binge eating, but behavioral weight loss is more effective for weight loss.

Medications for BED

Medications that have been tried for BED include antidepressants, appetite suppressants, and anticonvulsants.^25,26 Antidepressants are used to treat BED because:

• BED is often associated with depressive symptoms and disorders.
• BED is related to bulimia nervosa, and placebo-controlled trials have shown that the binge eating of bulimia nervosa responds to several classes of antidepressants. The selective serotonin reuptake inhibitor (SSRI) fluoxetine is the only medication indicated for treating any eating disorder (bulimia nervosa).
• Bupropion and venlafaxine—a serotonin-norepinephrine reuptake inhibitor (SNRI)—have weight-loss properties.

SSRIs are the most extensively studied antidepressants for treating BED. SSRIs have weightloss properties, but only short term.^25-26 Citalopram, fluoxetine, fluvoxamine, and sertraline have reduced binge eating and body weight more effectively than placebo during 6 to 9 weeks of treatment (Table 2).^25-26 However, one controlled study²³ showed that fluoxetine was not significantly more effective than placebo in reducing binge frequency or body weight after 16 weeks.

TCAs. Studies of tricyclic antidepressants (TCAs) for BED are sparse, and results have been mixed. In one trial, imipramine was similar to placebo in reducing binge frequency and body weight. In a placebo-controlled study of patients with nonpurging bulimia nervosa, desipramine reduced binge eating but had no effect on body weight.^25,26

Table 2

Drug therapies shown to be effective for BED*

Venlafaxine. In a retrospective review of 35 consecutive obese women with BED, venlafaxine, mean 222 mg/d for 28 to 300 days (median 120 days), reduced binge eating, body weight, and depressive symptoms.²⁷

Bupropion has been more effective than placebo for treating:

• uncomplicated obesity (short- and long-term)
• obesity associated with depressive symptoms
• bulimia nervosa (although bupropion is contraindicated in these patients because of seizure risk).^26,28,29

No controlled trials have studied bupropion for BED. When using dosages effective in depressive disorders, we find bupropion helpful in reducing binge eating, body weight, and depressive symptoms in BED patients.

Appetite suppressants decrease appetite and weight, may increase satiety, and may reduce depressive symptoms.

Sibutramine—a serotonin, norepinephrine, and dopamine reuptake inhibitor indicated for managing obesity—has been reported effective in BED in a 12-week, randomized, double-blind, placebo-controlled trial. A 15-mg/d dosage reduced binge frequency, body weight, and depressive symptoms more effectively than placebo in 60 obese patients with BED.³⁰ Most-frequent adverse effects (dry mouth and constipation) were mild and benign, and no significant complications were observed.

Sibutramine’s mechanism of action in BED is unknown. However, it suppressed food intake during binge-eating episodes in patients with BED in a randomized, controlled, cross-over laboratory study.³¹

Orlistat. We know of no published controlled studies of the lipase inhibitor orlistat in treating BED. In our experience, some patients do well with this agent, though we have observed infrequent purging episodes with it in patients with BED.

With orlistat, 120 mg tid, our BED patients have experienced weight loss comparable to that seen in uncomplicated obesity at similar dosages. Orlistat seems most effective for:

• patients whose binge eating is in remission
• those who responded to behavioral weightloss treatment, a psychological treatment, or another medication.

Anticonvulsants such as topiramate and zonisamide have been shown effective in treating obesity^32,33 and are sometimes used to treat BED. Obese BED patients with mood disorders often do best with psychotherapy plus medication

Topiramate at dosages of 50 to 600 mg/d (median 212 mg/d) reduced binge-eating frequency, obsessive-compulsive features of binge eating, and body weight more effectively than placebo in a 14-week study of 61 obese patients with BED. These effects were maintained across 48 weeks in an open-label extension trial.³⁴

Zonisamide, mean 513 mg/d, produced similar results during a prospective, open-label, 12-week trial in 15 patients with BED.³⁵ A controlled trial to replicate these findings is ongoing.

BED may respond to anticonvulsant therapy for several reasons:

• Some anticonvulsants are effective in treating bipolar disorder, which may occur with BED.¹²
• Some anticonvulsants have shown benefit in conditions associated with pathologic impulsivity, such as substance abuse, impulse-control, and cluster B disorders.¹⁰

Growing evidence shows that bulimia nervosa and BED may be associated with pathologic impulsivity.

Combination therapies are generally more effective than monotherapies in patients with mood disorders, uncomplicated obesity, and possibly bulimia nervosa. Even so, few trials have systematically studied combination therapy in managing patients with BED.

Two studies compared psychotherapy and antidepressants alone and in combination in treating BED.^21,23 Both showed that CBT alone was more effective in decreasing binge frequency than desipramine alone,²¹ fluoxetine alone,²³ and the combination of CBT and medication. On the other hand, patients who took desipramine either alone or in combination experienced a greater degree of weight loss than those who did not take desipramine.²¹

In another combination therapy, exercise has been shown to be an effective adjunct to CBT in maintenance treatment of obese women with BED.³⁶

No studies have compared behavioral weight management or a specialized psychotherapy in combination with an antiobesity drug or a weight-loss anticonvulsant in treating BED.

Treatment recommendations

In our experience, BED patients—particularly those with obesity and psychopathology—often do best with some combination of psychological treatment and medication:

• The psychological component may be behavioral weight-loss treatment, a specialized psychotherapy such as CBT or IPT, or some combination of behavioral weight-loss treatment and specialized psychotherapy.
• The medication component may consist of an antidepressant, anticonvulsant, antiobesity drug, or multiple drugs (such as an SSRI or sibutramine with topiramate for BED with major depression, or topiramate with lithium for BED with bipolar disorder).

Although combination therapies may be optimal for some patients, this approach remains unproven in controlled trials.

Patient preference. In addition to comorbidities, patient preference is an important consideration when choosing BED treatments. We determine our patients’ preferences by educating them as much as possible about their options. We explain the benefits and weaknesses of all treatments and encourage them to participate in forming their individualized treatment plans.

Patients sometimes have strong treatment preferences. Some prefer psychological treatments, whereas others prefer medications. Working with patient preferences enhances treatment adherence. For example, patients who fail a preferred treatment are often more willing to adhere to another treatment modality about which they initially were skeptical.

Related resources

• Bray GA, Bouchard C (eds). Handbook of obesity: clinical applications (2nd ed). New York, NY: Marcel Dekker, 2004.
• Cooper Z, Zairburn CG, Hawker DM. Cognitive behavioral treatment of obesity. A clinician’s guide. New York: Guilford Press, 2003.
• Carter WP, Hudson JI, Lalonde JK, et al. Pharmacologic treatment of binge eating disorder. Int J Eat Disord 2003;34(suppl):S74-88.

Drug brand names

• Bupropion • Wellbutrin
• Citalopram • Celexa
• Desipramine • Norpramin
• Fluoxetine • Prozac
• Fluvoxamine • Luvox
• Imipramine • Tofranil
• Lamotrigine • Lamictal
• Lithium • Eskalith, others
• Orlistat • Xenical
• Sertraline • Zoloft
• Sibutramine • Meridia
• Topiramate • Topamax
• Venlafaxine • Effexor
• Zonisamide • Zonegran

Disclosure

Dr. Kotwal receives grant support from Elan Corporation and is a speaker for Ortho-McNeil Pharmaceutical and Pfizer Inc.

Dr. Kaneria and Ms. Guerdjikova report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Dr. McElroy is a consultant to Abbott Laboratories, Bristol-Myers Squibb Co., Elan Corporation, GlaxoSmithKline, Janssen Pharmaceutica, Eli Lilly and Co., and Ortho-McNeil Pharmaceutical. She receives grant/research support from Elan Pharmaceuticals, Forest Pharmaceuticals, Merck & Co., Ortho-McNeil Pharmaceutical, and Sanofi-Synthelabo and is a speaker for Eli Lilly and Co. and Ortho-McNeil Pharmaceutical.

Ms. J started losing weight deliberately at age 14 while attending boarding school. She lost 25 lbs by jogging 6 miles per day, exercising another 2 hours, avoiding meat, abusing laxatives, and drinking large quantities of coffee.

She was referred to a school counselor because of her weight loss and returned home. She was happier at a local high school and recovered to normal weight. In college, however, she reverted to compulsive exercising and preoccupation with her weight after the break-up of her first intimate relationship.

Now at age 22, Ms. J has persistently failed to gain weight during outpatient therapy for anorexia nervosa. At 5′7″ she weighs 98 lbs. On the day she was to be hospitalized involuntarily, she took 25 diphenhydramine tablets, which her psychiatrist viewed as a suicide threat. The overdose was treated in the emergency room with ipecac syrup, and she was admitted for inpatient eating disorder treatment.

Like Ms. J, patients with anorexia nervosa resist treatment and deny having most diagnostic signs and symptoms. Based on the evidence and my 30 years of treating anorectic patients,¹ this article offers suggestions to help you:

• gather accurate histories from patients and their families
• identify common psychiatric comorbidities
• gain the patient’s trust during treatment
• provide effective dual therapy, with cognitive-behavioral and pharmacologic components.

Table 1

Diagnostic criteria for anorexia nervosa

Making the diagnosis

Anorexia nervosa is characterized by underweight, fear of gaining weight, disturbed body concept, and amenorrhea (Table 1). Its core psychological symptoms have been described as:

• relentless pursuit of thinness
• denial of cachexia
• and feelings of general ineffectiveness.²

The patient may say she feels fat even though emaciated or that parts of her body are too large. This disturbed experience of body weight or shape may represent sublimation and displacement for feelings of inadequacy. Because anorectic patients stay thin so effectively, they may feel a sense of accomplishment by evaluating themselves in terms of their thinness. Cognitive therapy focuses on correcting patients’ pervasive sense of inadequacy, as manifest in maturity fears and lack of confidence in coping with life’s problems.³

Subtypes. Anorexia nervosa has two subtypes—restricting and binge eating/purging—that differ in behavioral and medical symptoms.⁴ Patients with binge eating/purging show:

• higher rates of impulsivity (suicide attempts, self-mutilation, stealing, and alcohol and other substance abuse)
• more-prevalent impulsive personality disorders (borderline personality disorder, hysterical personality disorder)
• medical problems caused by purging.

Restricting-type patients are often dependent and submissive, with difficulty separating from parents. These patients may be preoccupied with orderliness, perfectionism, and control.

Recommendation. A structured interview to diagnose anorexia nervosa is summarized in Table 2. Because the patient will likely deny her symptoms, it is usually necessary to also interview family members or close friends.

Psychiatric comorbidity

Case report continued: A ‘perfectionist.’

School for Ms. J required great effort, and she spent many hours studying. Her upper-middle-class parents described her as “a perfectionist.” The family placed considerable emphasis on doing the “correct” thing.

During adolescence, Ms. J developed a major depressive episode that lasted 4 months. She also developed obsessions and compulsions unrelated to her eating disorder. She obsessively ruminated about the correct things to say in social circumstances and devoted 4 hours per day to cleaning and checking compulsions. She felt she had to wash her car every time before going out; if she could not, she would cancel her social plans.

Table 2

Diagnosis of anorexia nervosa: Questions to ask*

In college, she began abusing alcohol and was arrested once for driving while intoxicated.

Depression is the most common comorbidity in anorexia nervosa. Two-thirds of anorectic patients in a 10-year follow-up study reported a history of major depressive disorder.⁵ Suicide, starvation, and electrolyte imbalance are the three major causes of death. Among severely ill patients who require hospitalization, 10% to 20% die, though the suicide rate is undocumented.

Compulsions. Anorectics’ preoccupations about food and eating rituals have been compared with compulsions, though less than 20% of patients meet diagnostic criteria for obsessive-compulsive disorder.⁶

Substance abuse. Bulimic anorectics report more alcohol and substance use and abuse than restricting anorectics.⁷ The most common substances of abuse are cannabis, cocaine, stimulants, and over-the-counter pills such as diet aids.

Personality disorders. Up to 50% of patients with anorexia nervosa—particularly the binge/purge subtype—have personality disorders. Borderline personality disorder is especially common among binge/purge types,⁸ and avoidant personality disorder is more common among restricting types.

Table 3

Diagnostic signs of emaciation and purging in patients with anorexia nervosa

Personality disorders usually reflect instability in interpersonal relationships, poor self-image, or fluctuating affect. Patients may show a pattern of social inhibition, feelings of inadequacy, and hypersensitivity to negative evaluation.

Sexuality. Psychosocial and sexual development is often delayed in adolescent anorectics. In adults, interest in sex often plummets with anorexia onset, although binge/purge-type patients occasionally become promiscuous.

Medical signs

Case report continued: Abnormal ECG.

Ms. J was hospitalized after her weight dropped below 75% of normal for her age, height, and body build. She showed signs of electrolyte disturbance, including severe bradycardia (pulse rate 40) and ST-segment abnormalities on ECG.

Clinical signs of emaciation and purging can assist with diagnosis and in making decisions about medical treatment, including hospitalization (Table 3). Patients who purge are often weak and have puffy cheeks or parotid gland enlargement. They may have fainting spells and scars on their hands from stimulating vomiting. Laxative abuse may decrease colon motility and worsen constipation.

Neuroendocrine changes secondary to dieting and weight loss include:

• increased corticotropin-releasing hormone secretion
• blunted diurnal cortisol fluctuation
• decreased follicle-stimulating hormone (FSH) secretion
• impaired growth hormone regulation
• decreased luteinizing hormone (LH) secretion
• mildly decreased triiodothyronine
• erratic vasopressin secretion.

Measuring these changes is unnecessary, as general nutritional rehabilitation with weight gain will correct them.

Neurotransmitter function. Emaciated anorectics have a blunted response to pharmacologic probes for dopamine, reduced CSF norepinephrine turnover, and decreased CSF serotonin. Neuroimaging studies suggest that serotonin dysfunction may persist after weight is restored, although these findings require replication.

Treatment priorities

Effective therapies. Open studies indicate that multidimensional treatment—medical management, psychoeducation, and individual cognitive-behavioral therapy (CBT)—is most effective for anorexia nervosa. The fewer than 10 controlled trials that address anorexia nervosa treatment show:

• the more severe the illness, the more intense the treatment required
• outpatient therapy is most successful in patients who have had the illness <6 months, are not binging and vomiting, and have parents who participate in family therapy.

Hospitalization. An emaciated patient who is irritable, depressed, preoccupied with food, and sleep-deprived is unlikely to make progress toward behavioral change. The first goal, therefore, is to restore her nutritional state to normal.

Severely ill anorectic patients require hospitalization for daily monitoring of weight, calorie intake, urine output, and serum electrolytes and amylase (to assess purging behavior). Hospitalization is indicated for:

• loss of >20% of normal weight for age, height, and bone structure
• >6 months of repeated hospitalizations and underweight
• psychotic depression or serious suicide attempt
• incapacitating obsessions and compulsions, related or not to the eating disorder
• serious comorbid medical conditions, such as edema, hypoproteinemia, severe anemia, cardiac arrhythmia, or hypokalemic alkalosis (serum K+ < 2.5 mEq/L).

Keeping a patient in the hospital long enough to provide effective medical and psychological therapy has become difficult, however, because of medical insurance restrictions (Box). The result: poorer outcomes and increased relapse rates compared with 10 years ago.^9-12

Box

Nutritional rehabilitation and behavior changes can often correct the medical complications of emaciation and purging. Lost bone density is seldom restored, but nutritional rehabilitation can prevent further bone loss.¹³ Women who remain amenorrheic for several years after weight restoration tend to be more psychologically disturbed than those who resume menses rapidly.¹⁴

Cognitive-behavioral therapy

Other authors have discussed CBT for anorexia nervosa.^3,15 In general, the key tasks—operationalizing beliefs, evaluating autonomic thoughts, testing prospective hypotheses, and examining underlying assumptions—are accomplished by assessing anorexia’s distorted cognitions. No satisfactory controlled studies have examined any other type of individual psychotherapy for treating anorexia nervosa.

Alliance building. Patients with anorexia find it difficult to participate in therapeutic relationships. They are terrified of gaining weight and readily drop out of treatment. To build a therapeutic alliance:

• begin by helping the patient develop a history of her significant life events
• proceed slowly, praising her for every small attempt at changing her behavior
• set realistic therapy goals, considering her degree of resistance.

Monitoring. Behavior therapy consists primarily of positive reinforcements for weight gain. For this, we weigh outpatients weekly and inpatients daily. Outpatients are taught to keep diaries of daily food intake, stressful events, and emotional responses to them. The therapist begins each session by examining the patient’s diary with her and discussing how life events affect her eating behavior.

Cognitive restructuring helps patients identify their disturbed cognitions and challenge core beliefs about self-image. In this process, they become aware of their negative thoughts and develop arguments and evidence to support and refute the thoughts’ validity. They then form a reasoned conclusion based on the evidence.

Even if patients do not accept this logical conclusion, we encourage them to behave as if they believe it to be true. By doing this repeatedly, they eventually obtain some symptom relief.

Response-prevention techniques can help stop binging and purging. For example, we may require inpatients to sit together for 1 hour after eating. Because most patients will not vomit in front of each other, they learn how to resist vomiting and eventually experience reduced anxiety without vomiting after a meal.

Problem solving helps patients to reason through difficult food-related or interpersonal situations. The patient states the problem, then generates as many solutions as possible with the therapist’s assistance. She chooses one solution and puts it into effect, usually for 1 week. She then discusses the results with her therapist and decides whether to try another solution.

Family therapy. A family analysis—including a brief psychiatric history and evaluation of interactions—is recommended for all patients who live at home. This analysis can help you decide what type of family therapy or counseling to recommend.

Some families respond well with the parents and patient together in therapy sessions, whereas others are more comfortable with separate counseling. In a recent controlled study, anorectic patients younger than age 18 did equally well whether they were counseled with the family or separately.¹⁶

Brief therapy sessions are sometimes the most effective method to address family issues. When this is not possible, you and the patient can discuss family relationships in individual therapy.

Medications

Many medications have been used to treat anorexia nervosa, though few randomized, placebo-controlled studies exist. Because evidence does not support using psychotropics as monotherapy for anorexia nervosa, medication is considered adjunctive to CBT.

Chlorpromazine can help the hospitalized, severely ill patient who is overwhelmed with uncontrollable behavioral rituals and thoughts of losing weight. This antipsychotic helps reduce anorectic preoccupations and anxiety and helps make patients more amenable to therapy.

Start chlorpromazine at 10 mg tid and increase gradually until the patient can eat without extreme anxiety. Usual maximum dosage is 50 mg tid. Monitor blood pressure, tardive dyskinesia, and decreased white blood cell count.

Olanzapine may help induce weight gain and reduce anxiety in anorectic patients.¹⁷ Controlled and open-label studies are under way.

We start olanzapine at 2.5 mg/d and increase gradually to 10 or 15 mg/d. At this dosage, patients’ anxiety about eating is usually substantially reduced. Sedation is the most common side effect.

Anorexia patients often refuse to take olanzapine for fear of weight gain. If a patient’s emaciation is life-threatening, we may seek court permission to medicate her involuntarily. We reassure her that we will discontinue olanzapine when she reaches her target weight.

Serotonin in anorexia. Central serotonin pathways modulate feeding behavior. Serotonin antagonists—such as cyproheptadine—increase food intake and weight gain, whereas serotonin agonists—such as selective serotonin reuptake inhibitors (SSRIs)—decrease food intake.

Serotonin pathways also may modulate obsessive-compulsive and impulsive behaviors. Both serotonin agonists and antagonists can be useful adjuncts in treating anorexia nervosa.

In a double-blind, placebo-controlled trial, cyproheptadine, 4 to 8 mg tid, was associated with weight gain and reduced depressive symptoms in anorexia nervosa patients.¹⁸ Unlike tricyclic antidepressants, cyproheptadine does not reduce blood pressure or increase heart rate, which makes it attractive for emaciated anorectic patients. Dosages up to 28 mg/d can be used safely.

The SSRI fluoxetine may help prevent weight loss relapse in anorexia nervosa and reduce obsessive-compulsive behaviors.¹⁹ In open studies of low-weight anorectics, however, fluoxetine had little impact on weight or other clinically meaningful variables.²⁰ Thus, this agent is recommended for preventing weight-loss relapse only in patients who are within 10% to 15% of ideal body weight.

Outpatient care

Case report continued: Ongoing therapy

During hospitalization, Ms. J participated in all therapeutic modalities but had difficulty eating enough to gain weight. She reached her target weight of 127 lbs in about 7 weeks but gained no sense of purpose in life.

She is starting an intensive outpatient program using CBT to maintain her weight and further address the core psychopathology of her illness. Her maintenance therapy includes attending Alcoholics Anonymous meetings, ongoing fluoxetine (20 mg/d) to prevent weight-loss relapse, and CBT for obsessions and compulsions not related to her eating disorder.

Related resources

• Halmi KA. Eating disorders: Anorexia, bulimia nervosa and obesity. In: Hales RE, Yudofsky SC (eds). Textbook of clinical psychiatry, 4th ed. Washington, DC: American Psychiatric Publishing, 2003:1001-21.
• Academy of Eating Disorders. www.aedweb.org
• Anorexia Nervosa & Associated Disorders (ANAD). www.anad.org

Drug brand names

• Chlorpromazine • Thorazine
• Cyproheptadine • Periactin
• Fluoxetine • Prozac
• Olanzapine • Zyprexa

Disclosure

The author reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Ms. J started losing weight deliberately at age 14 while attending boarding school. She lost 25 lbs by jogging 6 miles per day, exercising another 2 hours, avoiding meat, abusing laxatives, and drinking large quantities of coffee.

She was referred to a school counselor because of her weight loss and returned home. She was happier at a local high school and recovered to normal weight. In college, however, she reverted to compulsive exercising and preoccupation with her weight after the break-up of her first intimate relationship.

Now at age 22, Ms. J has persistently failed to gain weight during outpatient therapy for anorexia nervosa. At 5′7″ she weighs 98 lbs. On the day she was to be hospitalized involuntarily, she took 25 diphenhydramine tablets, which her psychiatrist viewed as a suicide threat. The overdose was treated in the emergency room with ipecac syrup, and she was admitted for inpatient eating disorder treatment.

Like Ms. J, patients with anorexia nervosa resist treatment and deny having most diagnostic signs and symptoms. Based on the evidence and my 30 years of treating anorectic patients,¹ this article offers suggestions to help you:

• gather accurate histories from patients and their families
• identify common psychiatric comorbidities
• gain the patient’s trust during treatment
• provide effective dual therapy, with cognitive-behavioral and pharmacologic components.

Table 1

Diagnostic criteria for anorexia nervosa

Making the diagnosis

Anorexia nervosa is characterized by underweight, fear of gaining weight, disturbed body concept, and amenorrhea (Table 1). Its core psychological symptoms have been described as:

• relentless pursuit of thinness
• denial of cachexia
• and feelings of general ineffectiveness.²

The patient may say she feels fat even though emaciated or that parts of her body are too large. This disturbed experience of body weight or shape may represent sublimation and displacement for feelings of inadequacy. Because anorectic patients stay thin so effectively, they may feel a sense of accomplishment by evaluating themselves in terms of their thinness. Cognitive therapy focuses on correcting patients’ pervasive sense of inadequacy, as manifest in maturity fears and lack of confidence in coping with life’s problems.³

Subtypes. Anorexia nervosa has two subtypes—restricting and binge eating/purging—that differ in behavioral and medical symptoms.⁴ Patients with binge eating/purging show:

• higher rates of impulsivity (suicide attempts, self-mutilation, stealing, and alcohol and other substance abuse)
• more-prevalent impulsive personality disorders (borderline personality disorder, hysterical personality disorder)
• medical problems caused by purging.

Restricting-type patients are often dependent and submissive, with difficulty separating from parents. These patients may be preoccupied with orderliness, perfectionism, and control.

Recommendation. A structured interview to diagnose anorexia nervosa is summarized in Table 2. Because the patient will likely deny her symptoms, it is usually necessary to also interview family members or close friends.

Psychiatric comorbidity

Case report continued: A ‘perfectionist.’

School for Ms. J required great effort, and she spent many hours studying. Her upper-middle-class parents described her as “a perfectionist.” The family placed considerable emphasis on doing the “correct” thing.

During adolescence, Ms. J developed a major depressive episode that lasted 4 months. She also developed obsessions and compulsions unrelated to her eating disorder. She obsessively ruminated about the correct things to say in social circumstances and devoted 4 hours per day to cleaning and checking compulsions. She felt she had to wash her car every time before going out; if she could not, she would cancel her social plans.

Table 2

Diagnosis of anorexia nervosa: Questions to ask*

In college, she began abusing alcohol and was arrested once for driving while intoxicated.

Depression is the most common comorbidity in anorexia nervosa. Two-thirds of anorectic patients in a 10-year follow-up study reported a history of major depressive disorder.⁵ Suicide, starvation, and electrolyte imbalance are the three major causes of death. Among severely ill patients who require hospitalization, 10% to 20% die, though the suicide rate is undocumented.

Compulsions. Anorectics’ preoccupations about food and eating rituals have been compared with compulsions, though less than 20% of patients meet diagnostic criteria for obsessive-compulsive disorder.⁶

Substance abuse. Bulimic anorectics report more alcohol and substance use and abuse than restricting anorectics.⁷ The most common substances of abuse are cannabis, cocaine, stimulants, and over-the-counter pills such as diet aids.

Personality disorders. Up to 50% of patients with anorexia nervosa—particularly the binge/purge subtype—have personality disorders. Borderline personality disorder is especially common among binge/purge types,⁸ and avoidant personality disorder is more common among restricting types.

Table 3

Diagnostic signs of emaciation and purging in patients with anorexia nervosa

Personality disorders usually reflect instability in interpersonal relationships, poor self-image, or fluctuating affect. Patients may show a pattern of social inhibition, feelings of inadequacy, and hypersensitivity to negative evaluation.

Sexuality. Psychosocial and sexual development is often delayed in adolescent anorectics. In adults, interest in sex often plummets with anorexia onset, although binge/purge-type patients occasionally become promiscuous.

Medical signs

Case report continued: Abnormal ECG.

Ms. J was hospitalized after her weight dropped below 75% of normal for her age, height, and body build. She showed signs of electrolyte disturbance, including severe bradycardia (pulse rate 40) and ST-segment abnormalities on ECG.

Clinical signs of emaciation and purging can assist with diagnosis and in making decisions about medical treatment, including hospitalization (Table 3). Patients who purge are often weak and have puffy cheeks or parotid gland enlargement. They may have fainting spells and scars on their hands from stimulating vomiting. Laxative abuse may decrease colon motility and worsen constipation.

Neuroendocrine changes secondary to dieting and weight loss include:

• increased corticotropin-releasing hormone secretion
• blunted diurnal cortisol fluctuation
• decreased follicle-stimulating hormone (FSH) secretion
• impaired growth hormone regulation
• decreased luteinizing hormone (LH) secretion
• mildly decreased triiodothyronine
• erratic vasopressin secretion.

Measuring these changes is unnecessary, as general nutritional rehabilitation with weight gain will correct them.

Neurotransmitter function. Emaciated anorectics have a blunted response to pharmacologic probes for dopamine, reduced CSF norepinephrine turnover, and decreased CSF serotonin. Neuroimaging studies suggest that serotonin dysfunction may persist after weight is restored, although these findings require replication.

Treatment priorities

Effective therapies. Open studies indicate that multidimensional treatment—medical management, psychoeducation, and individual cognitive-behavioral therapy (CBT)—is most effective for anorexia nervosa. The fewer than 10 controlled trials that address anorexia nervosa treatment show:

• the more severe the illness, the more intense the treatment required
• outpatient therapy is most successful in patients who have had the illness <6 months, are not binging and vomiting, and have parents who participate in family therapy.

Hospitalization. An emaciated patient who is irritable, depressed, preoccupied with food, and sleep-deprived is unlikely to make progress toward behavioral change. The first goal, therefore, is to restore her nutritional state to normal.

Severely ill anorectic patients require hospitalization for daily monitoring of weight, calorie intake, urine output, and serum electrolytes and amylase (to assess purging behavior). Hospitalization is indicated for:

• loss of >20% of normal weight for age, height, and bone structure
• >6 months of repeated hospitalizations and underweight
• psychotic depression or serious suicide attempt
• incapacitating obsessions and compulsions, related or not to the eating disorder
• serious comorbid medical conditions, such as edema, hypoproteinemia, severe anemia, cardiac arrhythmia, or hypokalemic alkalosis (serum K+ < 2.5 mEq/L).

Keeping a patient in the hospital long enough to provide effective medical and psychological therapy has become difficult, however, because of medical insurance restrictions (Box). The result: poorer outcomes and increased relapse rates compared with 10 years ago.^9-12

Box

Nutritional rehabilitation and behavior changes can often correct the medical complications of emaciation and purging. Lost bone density is seldom restored, but nutritional rehabilitation can prevent further bone loss.¹³ Women who remain amenorrheic for several years after weight restoration tend to be more psychologically disturbed than those who resume menses rapidly.¹⁴

Cognitive-behavioral therapy

Other authors have discussed CBT for anorexia nervosa.^3,15 In general, the key tasks—operationalizing beliefs, evaluating autonomic thoughts, testing prospective hypotheses, and examining underlying assumptions—are accomplished by assessing anorexia’s distorted cognitions. No satisfactory controlled studies have examined any other type of individual psychotherapy for treating anorexia nervosa.

Alliance building. Patients with anorexia find it difficult to participate in therapeutic relationships. They are terrified of gaining weight and readily drop out of treatment. To build a therapeutic alliance:

• begin by helping the patient develop a history of her significant life events
• proceed slowly, praising her for every small attempt at changing her behavior
• set realistic therapy goals, considering her degree of resistance.

Monitoring. Behavior therapy consists primarily of positive reinforcements for weight gain. For this, we weigh outpatients weekly and inpatients daily. Outpatients are taught to keep diaries of daily food intake, stressful events, and emotional responses to them. The therapist begins each session by examining the patient’s diary with her and discussing how life events affect her eating behavior.

Cognitive restructuring helps patients identify their disturbed cognitions and challenge core beliefs about self-image. In this process, they become aware of their negative thoughts and develop arguments and evidence to support and refute the thoughts’ validity. They then form a reasoned conclusion based on the evidence.

Even if patients do not accept this logical conclusion, we encourage them to behave as if they believe it to be true. By doing this repeatedly, they eventually obtain some symptom relief.

Response-prevention techniques can help stop binging and purging. For example, we may require inpatients to sit together for 1 hour after eating. Because most patients will not vomit in front of each other, they learn how to resist vomiting and eventually experience reduced anxiety without vomiting after a meal.

Problem solving helps patients to reason through difficult food-related or interpersonal situations. The patient states the problem, then generates as many solutions as possible with the therapist’s assistance. She chooses one solution and puts it into effect, usually for 1 week. She then discusses the results with her therapist and decides whether to try another solution.

Family therapy. A family analysis—including a brief psychiatric history and evaluation of interactions—is recommended for all patients who live at home. This analysis can help you decide what type of family therapy or counseling to recommend.

Some families respond well with the parents and patient together in therapy sessions, whereas others are more comfortable with separate counseling. In a recent controlled study, anorectic patients younger than age 18 did equally well whether they were counseled with the family or separately.¹⁶

Brief therapy sessions are sometimes the most effective method to address family issues. When this is not possible, you and the patient can discuss family relationships in individual therapy.

Medications

Many medications have been used to treat anorexia nervosa, though few randomized, placebo-controlled studies exist. Because evidence does not support using psychotropics as monotherapy for anorexia nervosa, medication is considered adjunctive to CBT.

Chlorpromazine can help the hospitalized, severely ill patient who is overwhelmed with uncontrollable behavioral rituals and thoughts of losing weight. This antipsychotic helps reduce anorectic preoccupations and anxiety and helps make patients more amenable to therapy.

Start chlorpromazine at 10 mg tid and increase gradually until the patient can eat without extreme anxiety. Usual maximum dosage is 50 mg tid. Monitor blood pressure, tardive dyskinesia, and decreased white blood cell count.

Olanzapine may help induce weight gain and reduce anxiety in anorectic patients.¹⁷ Controlled and open-label studies are under way.

We start olanzapine at 2.5 mg/d and increase gradually to 10 or 15 mg/d. At this dosage, patients’ anxiety about eating is usually substantially reduced. Sedation is the most common side effect.

Anorexia patients often refuse to take olanzapine for fear of weight gain. If a patient’s emaciation is life-threatening, we may seek court permission to medicate her involuntarily. We reassure her that we will discontinue olanzapine when she reaches her target weight.

Serotonin in anorexia. Central serotonin pathways modulate feeding behavior. Serotonin antagonists—such as cyproheptadine—increase food intake and weight gain, whereas serotonin agonists—such as selective serotonin reuptake inhibitors (SSRIs)—decrease food intake.

Serotonin pathways also may modulate obsessive-compulsive and impulsive behaviors. Both serotonin agonists and antagonists can be useful adjuncts in treating anorexia nervosa.

In a double-blind, placebo-controlled trial, cyproheptadine, 4 to 8 mg tid, was associated with weight gain and reduced depressive symptoms in anorexia nervosa patients.¹⁸ Unlike tricyclic antidepressants, cyproheptadine does not reduce blood pressure or increase heart rate, which makes it attractive for emaciated anorectic patients. Dosages up to 28 mg/d can be used safely.

The SSRI fluoxetine may help prevent weight loss relapse in anorexia nervosa and reduce obsessive-compulsive behaviors.¹⁹ In open studies of low-weight anorectics, however, fluoxetine had little impact on weight or other clinically meaningful variables.²⁰ Thus, this agent is recommended for preventing weight-loss relapse only in patients who are within 10% to 15% of ideal body weight.

Outpatient care

Case report continued: Ongoing therapy

During hospitalization, Ms. J participated in all therapeutic modalities but had difficulty eating enough to gain weight. She reached her target weight of 127 lbs in about 7 weeks but gained no sense of purpose in life.

She is starting an intensive outpatient program using CBT to maintain her weight and further address the core psychopathology of her illness. Her maintenance therapy includes attending Alcoholics Anonymous meetings, ongoing fluoxetine (20 mg/d) to prevent weight-loss relapse, and CBT for obsessions and compulsions not related to her eating disorder.

Related resources

• Halmi KA. Eating disorders: Anorexia, bulimia nervosa and obesity. In: Hales RE, Yudofsky SC (eds). Textbook of clinical psychiatry, 4th ed. Washington, DC: American Psychiatric Publishing, 2003:1001-21.
• Academy of Eating Disorders. www.aedweb.org
• Anorexia Nervosa & Associated Disorders (ANAD). www.anad.org

Drug brand names

• Chlorpromazine • Thorazine
• Cyproheptadine • Periactin
• Fluoxetine • Prozac
• Olanzapine • Zyprexa

Disclosure

The author reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Ms. J started losing weight deliberately at age 14 while attending boarding school. She lost 25 lbs by jogging 6 miles per day, exercising another 2 hours, avoiding meat, abusing laxatives, and drinking large quantities of coffee.

She was referred to a school counselor because of her weight loss and returned home. She was happier at a local high school and recovered to normal weight. In college, however, she reverted to compulsive exercising and preoccupation with her weight after the break-up of her first intimate relationship.

Now at age 22, Ms. J has persistently failed to gain weight during outpatient therapy for anorexia nervosa. At 5′7″ she weighs 98 lbs. On the day she was to be hospitalized involuntarily, she took 25 diphenhydramine tablets, which her psychiatrist viewed as a suicide threat. The overdose was treated in the emergency room with ipecac syrup, and she was admitted for inpatient eating disorder treatment.

Like Ms. J, patients with anorexia nervosa resist treatment and deny having most diagnostic signs and symptoms. Based on the evidence and my 30 years of treating anorectic patients,¹ this article offers suggestions to help you:

• gather accurate histories from patients and their families
• identify common psychiatric comorbidities
• gain the patient’s trust during treatment
• provide effective dual therapy, with cognitive-behavioral and pharmacologic components.

Table 1

Diagnostic criteria for anorexia nervosa

Making the diagnosis

Anorexia nervosa is characterized by underweight, fear of gaining weight, disturbed body concept, and amenorrhea (Table 1). Its core psychological symptoms have been described as:

• relentless pursuit of thinness
• denial of cachexia
• and feelings of general ineffectiveness.²

The patient may say she feels fat even though emaciated or that parts of her body are too large. This disturbed experience of body weight or shape may represent sublimation and displacement for feelings of inadequacy. Because anorectic patients stay thin so effectively, they may feel a sense of accomplishment by evaluating themselves in terms of their thinness. Cognitive therapy focuses on correcting patients’ pervasive sense of inadequacy, as manifest in maturity fears and lack of confidence in coping with life’s problems.³

Subtypes. Anorexia nervosa has two subtypes—restricting and binge eating/purging—that differ in behavioral and medical symptoms.⁴ Patients with binge eating/purging show:

• higher rates of impulsivity (suicide attempts, self-mutilation, stealing, and alcohol and other substance abuse)
• more-prevalent impulsive personality disorders (borderline personality disorder, hysterical personality disorder)
• medical problems caused by purging.

Restricting-type patients are often dependent and submissive, with difficulty separating from parents. These patients may be preoccupied with orderliness, perfectionism, and control.

Recommendation. A structured interview to diagnose anorexia nervosa is summarized in Table 2. Because the patient will likely deny her symptoms, it is usually necessary to also interview family members or close friends.

Psychiatric comorbidity

Case report continued: A ‘perfectionist.’

School for Ms. J required great effort, and she spent many hours studying. Her upper-middle-class parents described her as “a perfectionist.” The family placed considerable emphasis on doing the “correct” thing.

During adolescence, Ms. J developed a major depressive episode that lasted 4 months. She also developed obsessions and compulsions unrelated to her eating disorder. She obsessively ruminated about the correct things to say in social circumstances and devoted 4 hours per day to cleaning and checking compulsions. She felt she had to wash her car every time before going out; if she could not, she would cancel her social plans.

Table 2

Diagnosis of anorexia nervosa: Questions to ask*

In college, she began abusing alcohol and was arrested once for driving while intoxicated.

Depression is the most common comorbidity in anorexia nervosa. Two-thirds of anorectic patients in a 10-year follow-up study reported a history of major depressive disorder.⁵ Suicide, starvation, and electrolyte imbalance are the three major causes of death. Among severely ill patients who require hospitalization, 10% to 20% die, though the suicide rate is undocumented.

Compulsions. Anorectics’ preoccupations about food and eating rituals have been compared with compulsions, though less than 20% of patients meet diagnostic criteria for obsessive-compulsive disorder.⁶

Substance abuse. Bulimic anorectics report more alcohol and substance use and abuse than restricting anorectics.⁷ The most common substances of abuse are cannabis, cocaine, stimulants, and over-the-counter pills such as diet aids.

Personality disorders. Up to 50% of patients with anorexia nervosa—particularly the binge/purge subtype—have personality disorders. Borderline personality disorder is especially common among binge/purge types,⁸ and avoidant personality disorder is more common among restricting types.

Table 3

Diagnostic signs of emaciation and purging in patients with anorexia nervosa

Personality disorders usually reflect instability in interpersonal relationships, poor self-image, or fluctuating affect. Patients may show a pattern of social inhibition, feelings of inadequacy, and hypersensitivity to negative evaluation.

Sexuality. Psychosocial and sexual development is often delayed in adolescent anorectics. In adults, interest in sex often plummets with anorexia onset, although binge/purge-type patients occasionally become promiscuous.

Medical signs

Case report continued: Abnormal ECG.

Ms. J was hospitalized after her weight dropped below 75% of normal for her age, height, and body build. She showed signs of electrolyte disturbance, including severe bradycardia (pulse rate 40) and ST-segment abnormalities on ECG.

Clinical signs of emaciation and purging can assist with diagnosis and in making decisions about medical treatment, including hospitalization (Table 3). Patients who purge are often weak and have puffy cheeks or parotid gland enlargement. They may have fainting spells and scars on their hands from stimulating vomiting. Laxative abuse may decrease colon motility and worsen constipation.

Neuroendocrine changes secondary to dieting and weight loss include:

• increased corticotropin-releasing hormone secretion
• blunted diurnal cortisol fluctuation
• decreased follicle-stimulating hormone (FSH) secretion
• impaired growth hormone regulation
• decreased luteinizing hormone (LH) secretion
• mildly decreased triiodothyronine
• erratic vasopressin secretion.

Measuring these changes is unnecessary, as general nutritional rehabilitation with weight gain will correct them.

Neurotransmitter function. Emaciated anorectics have a blunted response to pharmacologic probes for dopamine, reduced CSF norepinephrine turnover, and decreased CSF serotonin. Neuroimaging studies suggest that serotonin dysfunction may persist after weight is restored, although these findings require replication.

Treatment priorities

Effective therapies. Open studies indicate that multidimensional treatment—medical management, psychoeducation, and individual cognitive-behavioral therapy (CBT)—is most effective for anorexia nervosa. The fewer than 10 controlled trials that address anorexia nervosa treatment show:

• the more severe the illness, the more intense the treatment required
• outpatient therapy is most successful in patients who have had the illness <6 months, are not binging and vomiting, and have parents who participate in family therapy.

Hospitalization. An emaciated patient who is irritable, depressed, preoccupied with food, and sleep-deprived is unlikely to make progress toward behavioral change. The first goal, therefore, is to restore her nutritional state to normal.

Severely ill anorectic patients require hospitalization for daily monitoring of weight, calorie intake, urine output, and serum electrolytes and amylase (to assess purging behavior). Hospitalization is indicated for:

• loss of >20% of normal weight for age, height, and bone structure
• >6 months of repeated hospitalizations and underweight
• psychotic depression or serious suicide attempt
• incapacitating obsessions and compulsions, related or not to the eating disorder
• serious comorbid medical conditions, such as edema, hypoproteinemia, severe anemia, cardiac arrhythmia, or hypokalemic alkalosis (serum K+ < 2.5 mEq/L).

Keeping a patient in the hospital long enough to provide effective medical and psychological therapy has become difficult, however, because of medical insurance restrictions (Box). The result: poorer outcomes and increased relapse rates compared with 10 years ago.^9-12

Box

Nutritional rehabilitation and behavior changes can often correct the medical complications of emaciation and purging. Lost bone density is seldom restored, but nutritional rehabilitation can prevent further bone loss.¹³ Women who remain amenorrheic for several years after weight restoration tend to be more psychologically disturbed than those who resume menses rapidly.¹⁴

Cognitive-behavioral therapy

Other authors have discussed CBT for anorexia nervosa.^3,15 In general, the key tasks—operationalizing beliefs, evaluating autonomic thoughts, testing prospective hypotheses, and examining underlying assumptions—are accomplished by assessing anorexia’s distorted cognitions. No satisfactory controlled studies have examined any other type of individual psychotherapy for treating anorexia nervosa.

Alliance building. Patients with anorexia find it difficult to participate in therapeutic relationships. They are terrified of gaining weight and readily drop out of treatment. To build a therapeutic alliance:

• begin by helping the patient develop a history of her significant life events
• proceed slowly, praising her for every small attempt at changing her behavior
• set realistic therapy goals, considering her degree of resistance.

Monitoring. Behavior therapy consists primarily of positive reinforcements for weight gain. For this, we weigh outpatients weekly and inpatients daily. Outpatients are taught to keep diaries of daily food intake, stressful events, and emotional responses to them. The therapist begins each session by examining the patient’s diary with her and discussing how life events affect her eating behavior.

Cognitive restructuring helps patients identify their disturbed cognitions and challenge core beliefs about self-image. In this process, they become aware of their negative thoughts and develop arguments and evidence to support and refute the thoughts’ validity. They then form a reasoned conclusion based on the evidence.

Even if patients do not accept this logical conclusion, we encourage them to behave as if they believe it to be true. By doing this repeatedly, they eventually obtain some symptom relief.

Response-prevention techniques can help stop binging and purging. For example, we may require inpatients to sit together for 1 hour after eating. Because most patients will not vomit in front of each other, they learn how to resist vomiting and eventually experience reduced anxiety without vomiting after a meal.

Problem solving helps patients to reason through difficult food-related or interpersonal situations. The patient states the problem, then generates as many solutions as possible with the therapist’s assistance. She chooses one solution and puts it into effect, usually for 1 week. She then discusses the results with her therapist and decides whether to try another solution.

Family therapy. A family analysis—including a brief psychiatric history and evaluation of interactions—is recommended for all patients who live at home. This analysis can help you decide what type of family therapy or counseling to recommend.

Some families respond well with the parents and patient together in therapy sessions, whereas others are more comfortable with separate counseling. In a recent controlled study, anorectic patients younger than age 18 did equally well whether they were counseled with the family or separately.¹⁶

Brief therapy sessions are sometimes the most effective method to address family issues. When this is not possible, you and the patient can discuss family relationships in individual therapy.

Medications

Many medications have been used to treat anorexia nervosa, though few randomized, placebo-controlled studies exist. Because evidence does not support using psychotropics as monotherapy for anorexia nervosa, medication is considered adjunctive to CBT.

Chlorpromazine can help the hospitalized, severely ill patient who is overwhelmed with uncontrollable behavioral rituals and thoughts of losing weight. This antipsychotic helps reduce anorectic preoccupations and anxiety and helps make patients more amenable to therapy.

Start chlorpromazine at 10 mg tid and increase gradually until the patient can eat without extreme anxiety. Usual maximum dosage is 50 mg tid. Monitor blood pressure, tardive dyskinesia, and decreased white blood cell count.

Olanzapine may help induce weight gain and reduce anxiety in anorectic patients.¹⁷ Controlled and open-label studies are under way.

We start olanzapine at 2.5 mg/d and increase gradually to 10 or 15 mg/d. At this dosage, patients’ anxiety about eating is usually substantially reduced. Sedation is the most common side effect.

Anorexia patients often refuse to take olanzapine for fear of weight gain. If a patient’s emaciation is life-threatening, we may seek court permission to medicate her involuntarily. We reassure her that we will discontinue olanzapine when she reaches her target weight.

Serotonin in anorexia. Central serotonin pathways modulate feeding behavior. Serotonin antagonists—such as cyproheptadine—increase food intake and weight gain, whereas serotonin agonists—such as selective serotonin reuptake inhibitors (SSRIs)—decrease food intake.

Serotonin pathways also may modulate obsessive-compulsive and impulsive behaviors. Both serotonin agonists and antagonists can be useful adjuncts in treating anorexia nervosa.

In a double-blind, placebo-controlled trial, cyproheptadine, 4 to 8 mg tid, was associated with weight gain and reduced depressive symptoms in anorexia nervosa patients.¹⁸ Unlike tricyclic antidepressants, cyproheptadine does not reduce blood pressure or increase heart rate, which makes it attractive for emaciated anorectic patients. Dosages up to 28 mg/d can be used safely.

The SSRI fluoxetine may help prevent weight loss relapse in anorexia nervosa and reduce obsessive-compulsive behaviors.¹⁹ In open studies of low-weight anorectics, however, fluoxetine had little impact on weight or other clinically meaningful variables.²⁰ Thus, this agent is recommended for preventing weight-loss relapse only in patients who are within 10% to 15% of ideal body weight.

Outpatient care

Case report continued: Ongoing therapy

During hospitalization, Ms. J participated in all therapeutic modalities but had difficulty eating enough to gain weight. She reached her target weight of 127 lbs in about 7 weeks but gained no sense of purpose in life.

She is starting an intensive outpatient program using CBT to maintain her weight and further address the core psychopathology of her illness. Her maintenance therapy includes attending Alcoholics Anonymous meetings, ongoing fluoxetine (20 mg/d) to prevent weight-loss relapse, and CBT for obsessions and compulsions not related to her eating disorder.

Related resources

• Halmi KA. Eating disorders: Anorexia, bulimia nervosa and obesity. In: Hales RE, Yudofsky SC (eds). Textbook of clinical psychiatry, 4th ed. Washington, DC: American Psychiatric Publishing, 2003:1001-21.
• Academy of Eating Disorders. www.aedweb.org
• Anorexia Nervosa & Associated Disorders (ANAD). www.anad.org

Drug brand names

• Chlorpromazine • Thorazine
• Cyproheptadine • Periactin
• Fluoxetine • Prozac
• Olanzapine • Zyprexa

Disclosure

The author reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Compared with other psychiatric diagnoses, eating disorders are relatively new. Bulimia nervosa was first described as a distinct syndrome in 1979, ^{Update on Eating Disorders.” In part 1 Harrison G. Pope, Jr., MD, and James I. Hudson, MD, ScD, of Harvard Medical School describe how to avoid undertreating bulimia nervosa. Future issues will feature insights on:}

• anorexia nervosa by Katherine A. Halmi, MD, Weill Medical College of Cornell University
• binge eating by Susan L. McElroy, MD, Renu Kotwal, MD, and Rakesh M. Kaneria, MD, University of Cincinnati College of Medicine.

Rigorous clinical research by these experts and others has helped weed out unsubstantiated theories while providing empiric evidence of eating disorders’ biological and psychological roots. This series updates the evidence for choosing medications and psychotherapies that have shown the greatest therapeutic promise.

Compared with other psychiatric diagnoses, eating disorders are relatively new. Bulimia nervosa was first described as a distinct syndrome in 1979, ^{Update on Eating Disorders.” In part 1 Harrison G. Pope, Jr., MD, and James I. Hudson, MD, ScD, of Harvard Medical School describe how to avoid undertreating bulimia nervosa. Future issues will feature insights on:}

• anorexia nervosa by Katherine A. Halmi, MD, Weill Medical College of Cornell University
• binge eating by Susan L. McElroy, MD, Renu Kotwal, MD, and Rakesh M. Kaneria, MD, University of Cincinnati College of Medicine.

Rigorous clinical research by these experts and others has helped weed out unsubstantiated theories while providing empiric evidence of eating disorders’ biological and psychological roots. This series updates the evidence for choosing medications and psychotherapies that have shown the greatest therapeutic promise.

Compared with other psychiatric diagnoses, eating disorders are relatively new. Bulimia nervosa was first described as a distinct syndrome in 1979, ^{Update on Eating Disorders.” In part 1 Harrison G. Pope, Jr., MD, and James I. Hudson, MD, ScD, of Harvard Medical School describe how to avoid undertreating bulimia nervosa. Future issues will feature insights on:}

• anorexia nervosa by Katherine A. Halmi, MD, Weill Medical College of Cornell University
• binge eating by Susan L. McElroy, MD, Renu Kotwal, MD, and Rakesh M. Kaneria, MD, University of Cincinnati College of Medicine.

Rigorous clinical research by these experts and others has helped weed out unsubstantiated theories while providing empiric evidence of eating disorders’ biological and psychological roots. This series updates the evidence for choosing medications and psychotherapies that have shown the greatest therapeutic promise.

Thousands of scientific papers have been written about bulimia, but not all patients receive effective treatments that produce remission.

To set the record straight and help psychiatrists avoid undertreating bulimia, this article discusses:

• evidence for using antidepressants, even when patients are not “depressed”
• merits of psychotherapies, including those shown to work and those that can harm
• augmentation therapies that can help increase response from partial to full remission.

Initial evaluation

Diagnosis. Bulimia nervosa is characterized by eating binges, followed by purging behaviors such as self-induced vomiting or laxative abuse^1,2 (Table 1). It affects 1% to 3% of adolescent girls and young women and occurs in women 5 to 10 times more often than in men.

Bulimia is often persistent. About one-half of bulimic patients—including those who have been treated—continue to show eating disorder features on long-term follow-up.^3,4

Psychiatric comorbidity. Most bulimic patients report a history of other psychiatric disorders, especially major depressive and bipolar disorders and anxiety disorders such as panic disorder, social phobia, and obsessive-compulsive disorder (OCD).⁵ Because these psychiatric comorbidities may occur before, during, or after bulimia nervosa, one cannot assume that mood or anxiety disorders are a cause or consequence of bulimia. Instead, bulimia nervosa, mood disorders, and anxiety disorders may be different expressions of a shared etiologic abnormality.

Table 1

DSM-IV-TR diagnostic criteria for bulimia nervosa

Evidence supporting this hypothesis comes from studies showing that these disorders:

• respond to several chemically unrelated families of antidepressants^6,7
• frequently co-occur in individual patients^5,7
• frequently co-aggregate in families.^7-9

We have published this evidence^6,7 and proposed that bulimia nervosa may be one form of a larger underlying disorder, which we termed “affective spectrum disorder.”

Antidepressants are often rapidly effective in treating bulimic symptoms,¹⁰ regardless of whether patients exhibit depressive symptoms. Thus, there is no reason to withhold antidepressant therapy simply because a bulimic patient is not depressed. The term “antidepressant” may be a misnomer; these drugs are effective for numerous conditions, of which depression is only one.

Anorexic symptoms. Co-occurring depressive or anxiety disorders in a bulimic patient will not greatly alter treatment. The antidepressants and psychotherapies typically used to treat bulimia are often equally effective for affective disorders. Co-occurring anorexia nervosa, however, is a more serious concern.

Bulimic patients often display a history of anorexia nervosa; in many cases, the patient develops anorexia nervosa as a teenager and then progresses to bulimia nervosa across several years. Her prognosis is much better if her weight normalizes with the shift to bulimia nervosa, than if her weight remains well below normal for her height. It is unclear why medications and psychotherapy are much less effective in bulimic patients with anorexic symptoms than in those with bulimia alone. Watch for further details on anorexia nervosa as this series continues in future issues of.

Medical considerations. Potential medical complications—mostly consequences of vomiting or laxative use—are important to consider when you assess a bulimic patient:¹

• The acid in vomitus may gradually erode tooth enamel, requiring dental consultation.
• Vomitus may inflame salivary gland ducts, though the swelling is usually benign.
• Frequent vomiting may result in hypokalemia and alkalosis, although aggressive medical treatment usually is not needed.

Ask about ipecac use. To induce vomiting, some patients may abuse ipecac syrup, which can cause cardiomyopathy.¹¹

Inpatient or outpatient? Unless the bulimic patient displays severe and medically dangerous anorexic symptoms, she can usually be treated as an outpatient. However, evaluate her carefully for suicidal ideation—which is not uncommon in bulimia nervosa—and consider inpatient treatment if necessary.

Medication vs. psychotherapy

The relative merits of medication versus psychotherapy in treating bulimia nervosa continue to be debated. The Cochrane Database of Systematic Reviews includes meta-analyses of both drug therapy¹² and psychotherapy¹³ for bulimia nervosa. The 2001 drug therapy review found that “the use of a single antidepressant agent was clinically effective,” with no one drug clearly superior to another. Notably, this review was published before recent findings on topiramate.

The corresponding 2002 review of psychotherapy concludes—somewhat more cautiously—that “there is a small body of evidence for the efficacy of cognitive-behavior therapy in bulimia nervosa and similar syndromes, but the quality of trials is very variable and sample sizes are often small.”

In bulimia nervosa and other psychiatric disorders, comparing psychotherapy with drug therapy is hazardous because several factors bias the comparison in favor of psychotherapy. These factors include an expectational effect, a responsibility effect, and differential generalizability of study results.

Expectational effect. Patients in clinical trials are aware that they are receiving psychotherapy and, presumably, that study investigators hope to demonstrate its efficacy. This might account for much of psychotherapy’s apparent effect, as even placebos can produce 30% to 50% improvement in bulimia.¹⁴

Responsibility effect. If a patient fails to improve in a drug study, she will conclude that the drug has failed. But if she fails to improve in a psychotherapy study, she may conclude that she has failed. Because psychological treatments generally require patients to work in therapy, the patient may feel partially responsible for the outcome. Thus, to avoid cognitive dissonance, she may consciously or unconsciously exaggerate her improvement, both in her own mind and when reporting to treaters.

Differential generalizability. Psychological study protocols, such as administering several months of a behavioral treatment, usually mimic clinical practice fairly well. This is not the case with drug study protocols.

No responsible clinician would inflexibly administer a single dosage of a single drug for a fixed period to every bulimic patient and then declare failure for all nonresponders, as is done in study protocols. In practice, the clinician can offer nonresponders augmentation strategies and additional drug trials. Thus, calculations of bulimia response rates in drug studies substantially understate response to drug therapy in clinical practice.

Table 2

How effective are medications in treating bulimia nervosa?

One also might note that psychological study findings have not “sold” well in the clinical practice marketplace. For example, in a recent survey of more than 220 bulimic women treated with psychotherapy, only 6.9% said they received a full course of cognitive behavioral therapy (CBT)¹⁴ —despite two decades of evidence of its efficacy. By contrast, untested, inefficacious, and possibly harmful psychotherapies for bulimia—including recovered-memory therapy—appear to be thriving.

Recommendation. Interpret with caution any head-to-head comparisons of psychological versus drug therapies—especially when clinical practice recommendations are made. Certain psychological therapies provided by specifically-trained individuals likely do help patients with bulimia nervosa. However, biases inherent to the studies may inflate psychological therapies’ efficacy when compared with that of drug therapy.

Therefore, for a psychiatrist who does not specialize in eating disorders to offer exclusively psychological therapy to a bulimic patient—while withholding or postponing drug therapy—may now be a questionable practice.

Choosing drug therapies

Although consensus is lacking on an optimal treatment trial sequence for bulimia nervosa, we suggest a rational approach based on the evidence and our experience (Algorithm).

First-line antidepressants. A selective serotonin reuptake inhibitor (SSRI) trial is usually the first choice (Table 2), and some data suggest that higher-than-usual dosages may be required. For example, in a large multicenter trial of fluoxetine in bulimia nervosa, 60 mg/d was considerably more effective than 20 mg/d for reducing binge eating behavior and vomiting frequency.¹⁵

Based on our observations, however, we believe that noncompliance or irregular compliance may account for this difference in response. Bulimic patients’ impulsive and obsessional behavior may keep them from taking their medications as prescribed. The higher fluoxetine dosage may therefore have been more effective simply because it ensured adequate plasma levels, even when patients missed or forgot multiple doses.

Augmenting agents. A first antidepressant trial rarely leads to complete remission of bulimic symptoms. This is not a serious concern, however, because many other options are available.

Liothyronine. Partial responders to SSRIs often become complete responders when we add a 10-day trial of liothyronine (T3), 25 μg/d. If this fails, we may try augmenting with lithium carbonate, although bulimic patients are often afraid of weight gain or lithium’s other side effects.

Topiramate. A newer augmentation strategy is to add the anticonvulsant topiramate. Used alone, topiramate demonstrated effectiveness for bulimia nervosa in one placebo-controlled, double-blind trial.¹⁶

Adding topiramate to an antidepressant regimen will likely reduce any remaining bulimic symptoms. In addition, topiramate often produces weight loss—a side effect that bulimic patients usually welcome. It remains unclear whether topiramate’s weight-loss effects might pose a hazard in patients with simultaneous bulimic and anorexic symptoms.

Other antidepressants. If the above strategies fail, other antidepressant options include venlafaxine, tricyclics, and monoamine oxidase inhibitors. Bupropion is not recommended in bulimia nervosa; one trial¹⁷ of this agent resulted in a much higher rate of grand mal seizures in bulimic patients than in patients taking bupropion for depression.

In bulimic patients with concomitant bipolar disorder, the anticonvulsants carbamazepine and valproate often reduce affective and bulimic symptoms. By contrast, the anticonvulsant phenytoin—once thought to be useful in bulimia nervosa¹⁰ —offers little benefit for either bulimic or affective symptoms.

Persistence is important when initial medication trials fail. One unblinded study followed 36 bulimic patients 9 to 19 months after they completed a controlled study with trazodone.¹⁸ Of the 26 patients who tried a second or third antidepressant, 17 (65%) achieved remission of bulimia on follow-up. Of the 10 patients who declined a second or third trial, only 1 (10%) attained remission.

Notably, these study results were obtained before the SSRIs and other newer antidepressants or topiramate became available. Cooperative patients using present-day medications might be able to achieve remission rates that exceed 65%.

Algorithm Proposed treatment approach to bulimia nervosa

How effective are psychotherapies in treating bulimia nervosa?

Psychotherapy

Cognitive-behavioral therapy. CBT—given either individually or in groups—is the most effective psychotherapy for bulimia (Table 3).¹⁹ CBT typically involves 3 to 6 months of helping the patient focus on her bulimic behaviors and on specific attitudes—such as unrealistic preoccupations with being “too fat”—that perpetuate the behaviors.

In practice, unfortunately, few bulimic patients are offered CBT, perhaps because few clinicians are trained in the specific approach used for bulimia nervosa.¹⁹ If you are not trained in using CBT for bulimia and do not have access to colleagues who offer this treatment, you may begin with medication plus simple behavioral treatments, such as:

• offering supportive therapy in the office
• referring patients to self-help groups for persons with eating disorders.

If this strategy fails, encourage patients to consider CBT—even if they must travel some distance to obtain it.

Other specialized psychotherapies. Dialectical behavior therapy and interpersonal psychotherapy appear to be effective in bulimia. Again, however, clinicians who lack training in these techniques or access to local experts may be unable to offer them. Psychodynamic therapy does not appear to offer greater benefit in bulimia nervosa than ordinary supportive counseling.

Dubious therapies. One psychodynamic approach—regrettably still practiced—is “recovered memory therapy.” Therapists who use it claim that childhood sexual abuse or other trauma can cause bulimic symptoms but patients have repressed the memory of these events.²⁰

No methodologically sound evidence has shown that childhood sexual abuse can cause bulimia nervosa years or decades later.²¹ Nor is there acceptable evidence that people can repress the memory of a traumatic experience.²² Therapists administering recovered memory therapy have been subjected to malpractice judgments totaling tens of millions of dollars from suits filed by patients who eventually realized that so-called “recovered” memories were false.²³

Another dubious therapy—eye movement desensitization and reprocessing (EMDR)—also may involve attempts to “recover” memories of putative traumatic events.²⁴ No methodologically sound evidence has shown that EMDR is effective in bulimic patients, and the technique’s theoretical basis is questionable.^24,25

Related resources

• Mental Health Net/Eating disorders. http://eatingdisorders.mentalhelp.net
• National Association of Anorexia Nervosa and Associated Disorders (ANAD). www.altrue.net/site/anadweb/
• Association for Advancement of Behavior Therapy. www.aabt.org

Drug brand names

• Bupropion • Wellbutrin
• Citalopram • Celexa
• Desipramine • Norpramin
• Fluoxetine • Prozac
• Lithium • Lithobid, Eskalith
• Nortriptyline • Pamelor, Aventyl
• Sertraline • Zoloft
• Topiramate • Topamax
• Trazodone • Desyrel
• Liothyronine • Cytomel
• Venlafaxine • Effexor

Disclosure

Dr. Pope receives research support from Ortho-McNeil Pharmaceuticals and is a consultant to Solvay Pharmaceuticals and Auxilium Pharmaceuticals.

Dr. Hudson receives research support from and is a consultant to Eli Lilly & Co. and Ortho-McNeil Pharmaceuticals.

Thousands of scientific papers have been written about bulimia, but not all patients receive effective treatments that produce remission.

To set the record straight and help psychiatrists avoid undertreating bulimia, this article discusses:

• evidence for using antidepressants, even when patients are not “depressed”
• merits of psychotherapies, including those shown to work and those that can harm
• augmentation therapies that can help increase response from partial to full remission.

Initial evaluation

Diagnosis. Bulimia nervosa is characterized by eating binges, followed by purging behaviors such as self-induced vomiting or laxative abuse^1,2 (Table 1). It affects 1% to 3% of adolescent girls and young women and occurs in women 5 to 10 times more often than in men.

Bulimia is often persistent. About one-half of bulimic patients—including those who have been treated—continue to show eating disorder features on long-term follow-up.^3,4

Psychiatric comorbidity. Most bulimic patients report a history of other psychiatric disorders, especially major depressive and bipolar disorders and anxiety disorders such as panic disorder, social phobia, and obsessive-compulsive disorder (OCD).⁵ Because these psychiatric comorbidities may occur before, during, or after bulimia nervosa, one cannot assume that mood or anxiety disorders are a cause or consequence of bulimia. Instead, bulimia nervosa, mood disorders, and anxiety disorders may be different expressions of a shared etiologic abnormality.

Table 1

DSM-IV-TR diagnostic criteria for bulimia nervosa

Evidence supporting this hypothesis comes from studies showing that these disorders:

• respond to several chemically unrelated families of antidepressants^6,7
• frequently co-occur in individual patients^5,7
• frequently co-aggregate in families.^7-9

We have published this evidence^6,7 and proposed that bulimia nervosa may be one form of a larger underlying disorder, which we termed “affective spectrum disorder.”

Antidepressants are often rapidly effective in treating bulimic symptoms,¹⁰ regardless of whether patients exhibit depressive symptoms. Thus, there is no reason to withhold antidepressant therapy simply because a bulimic patient is not depressed. The term “antidepressant” may be a misnomer; these drugs are effective for numerous conditions, of which depression is only one.

Anorexic symptoms. Co-occurring depressive or anxiety disorders in a bulimic patient will not greatly alter treatment. The antidepressants and psychotherapies typically used to treat bulimia are often equally effective for affective disorders. Co-occurring anorexia nervosa, however, is a more serious concern.

Bulimic patients often display a history of anorexia nervosa; in many cases, the patient develops anorexia nervosa as a teenager and then progresses to bulimia nervosa across several years. Her prognosis is much better if her weight normalizes with the shift to bulimia nervosa, than if her weight remains well below normal for her height. It is unclear why medications and psychotherapy are much less effective in bulimic patients with anorexic symptoms than in those with bulimia alone. Watch for further details on anorexia nervosa as this series continues in future issues of.

Medical considerations. Potential medical complications—mostly consequences of vomiting or laxative use—are important to consider when you assess a bulimic patient:¹

• The acid in vomitus may gradually erode tooth enamel, requiring dental consultation.
• Vomitus may inflame salivary gland ducts, though the swelling is usually benign.
• Frequent vomiting may result in hypokalemia and alkalosis, although aggressive medical treatment usually is not needed.

Ask about ipecac use. To induce vomiting, some patients may abuse ipecac syrup, which can cause cardiomyopathy.¹¹

Inpatient or outpatient? Unless the bulimic patient displays severe and medically dangerous anorexic symptoms, she can usually be treated as an outpatient. However, evaluate her carefully for suicidal ideation—which is not uncommon in bulimia nervosa—and consider inpatient treatment if necessary.

Medication vs. psychotherapy

The relative merits of medication versus psychotherapy in treating bulimia nervosa continue to be debated. The Cochrane Database of Systematic Reviews includes meta-analyses of both drug therapy¹² and psychotherapy¹³ for bulimia nervosa. The 2001 drug therapy review found that “the use of a single antidepressant agent was clinically effective,” with no one drug clearly superior to another. Notably, this review was published before recent findings on topiramate.

The corresponding 2002 review of psychotherapy concludes—somewhat more cautiously—that “there is a small body of evidence for the efficacy of cognitive-behavior therapy in bulimia nervosa and similar syndromes, but the quality of trials is very variable and sample sizes are often small.”

In bulimia nervosa and other psychiatric disorders, comparing psychotherapy with drug therapy is hazardous because several factors bias the comparison in favor of psychotherapy. These factors include an expectational effect, a responsibility effect, and differential generalizability of study results.

Expectational effect. Patients in clinical trials are aware that they are receiving psychotherapy and, presumably, that study investigators hope to demonstrate its efficacy. This might account for much of psychotherapy’s apparent effect, as even placebos can produce 30% to 50% improvement in bulimia.¹⁴

Responsibility effect. If a patient fails to improve in a drug study, she will conclude that the drug has failed. But if she fails to improve in a psychotherapy study, she may conclude that she has failed. Because psychological treatments generally require patients to work in therapy, the patient may feel partially responsible for the outcome. Thus, to avoid cognitive dissonance, she may consciously or unconsciously exaggerate her improvement, both in her own mind and when reporting to treaters.

Differential generalizability. Psychological study protocols, such as administering several months of a behavioral treatment, usually mimic clinical practice fairly well. This is not the case with drug study protocols.

No responsible clinician would inflexibly administer a single dosage of a single drug for a fixed period to every bulimic patient and then declare failure for all nonresponders, as is done in study protocols. In practice, the clinician can offer nonresponders augmentation strategies and additional drug trials. Thus, calculations of bulimia response rates in drug studies substantially understate response to drug therapy in clinical practice.

Table 2

How effective are medications in treating bulimia nervosa?

One also might note that psychological study findings have not “sold” well in the clinical practice marketplace. For example, in a recent survey of more than 220 bulimic women treated with psychotherapy, only 6.9% said they received a full course of cognitive behavioral therapy (CBT)¹⁴ —despite two decades of evidence of its efficacy. By contrast, untested, inefficacious, and possibly harmful psychotherapies for bulimia—including recovered-memory therapy—appear to be thriving.

Recommendation. Interpret with caution any head-to-head comparisons of psychological versus drug therapies—especially when clinical practice recommendations are made. Certain psychological therapies provided by specifically-trained individuals likely do help patients with bulimia nervosa. However, biases inherent to the studies may inflate psychological therapies’ efficacy when compared with that of drug therapy.

Therefore, for a psychiatrist who does not specialize in eating disorders to offer exclusively psychological therapy to a bulimic patient—while withholding or postponing drug therapy—may now be a questionable practice.

Choosing drug therapies

Although consensus is lacking on an optimal treatment trial sequence for bulimia nervosa, we suggest a rational approach based on the evidence and our experience (Algorithm).

First-line antidepressants. A selective serotonin reuptake inhibitor (SSRI) trial is usually the first choice (Table 2), and some data suggest that higher-than-usual dosages may be required. For example, in a large multicenter trial of fluoxetine in bulimia nervosa, 60 mg/d was considerably more effective than 20 mg/d for reducing binge eating behavior and vomiting frequency.¹⁵

Based on our observations, however, we believe that noncompliance or irregular compliance may account for this difference in response. Bulimic patients’ impulsive and obsessional behavior may keep them from taking their medications as prescribed. The higher fluoxetine dosage may therefore have been more effective simply because it ensured adequate plasma levels, even when patients missed or forgot multiple doses.

Augmenting agents. A first antidepressant trial rarely leads to complete remission of bulimic symptoms. This is not a serious concern, however, because many other options are available.

Liothyronine. Partial responders to SSRIs often become complete responders when we add a 10-day trial of liothyronine (T3), 25 μg/d. If this fails, we may try augmenting with lithium carbonate, although bulimic patients are often afraid of weight gain or lithium’s other side effects.

Topiramate. A newer augmentation strategy is to add the anticonvulsant topiramate. Used alone, topiramate demonstrated effectiveness for bulimia nervosa in one placebo-controlled, double-blind trial.¹⁶

Adding topiramate to an antidepressant regimen will likely reduce any remaining bulimic symptoms. In addition, topiramate often produces weight loss—a side effect that bulimic patients usually welcome. It remains unclear whether topiramate’s weight-loss effects might pose a hazard in patients with simultaneous bulimic and anorexic symptoms.

Other antidepressants. If the above strategies fail, other antidepressant options include venlafaxine, tricyclics, and monoamine oxidase inhibitors. Bupropion is not recommended in bulimia nervosa; one trial¹⁷ of this agent resulted in a much higher rate of grand mal seizures in bulimic patients than in patients taking bupropion for depression.

In bulimic patients with concomitant bipolar disorder, the anticonvulsants carbamazepine and valproate often reduce affective and bulimic symptoms. By contrast, the anticonvulsant phenytoin—once thought to be useful in bulimia nervosa¹⁰ —offers little benefit for either bulimic or affective symptoms.

Persistence is important when initial medication trials fail. One unblinded study followed 36 bulimic patients 9 to 19 months after they completed a controlled study with trazodone.¹⁸ Of the 26 patients who tried a second or third antidepressant, 17 (65%) achieved remission of bulimia on follow-up. Of the 10 patients who declined a second or third trial, only 1 (10%) attained remission.

Notably, these study results were obtained before the SSRIs and other newer antidepressants or topiramate became available. Cooperative patients using present-day medications might be able to achieve remission rates that exceed 65%.

Algorithm Proposed treatment approach to bulimia nervosa

How effective are psychotherapies in treating bulimia nervosa?

Psychotherapy

Cognitive-behavioral therapy. CBT—given either individually or in groups—is the most effective psychotherapy for bulimia (Table 3).¹⁹ CBT typically involves 3 to 6 months of helping the patient focus on her bulimic behaviors and on specific attitudes—such as unrealistic preoccupations with being “too fat”—that perpetuate the behaviors.

In practice, unfortunately, few bulimic patients are offered CBT, perhaps because few clinicians are trained in the specific approach used for bulimia nervosa.¹⁹ If you are not trained in using CBT for bulimia and do not have access to colleagues who offer this treatment, you may begin with medication plus simple behavioral treatments, such as:

• offering supportive therapy in the office
• referring patients to self-help groups for persons with eating disorders.

If this strategy fails, encourage patients to consider CBT—even if they must travel some distance to obtain it.

Other specialized psychotherapies. Dialectical behavior therapy and interpersonal psychotherapy appear to be effective in bulimia. Again, however, clinicians who lack training in these techniques or access to local experts may be unable to offer them. Psychodynamic therapy does not appear to offer greater benefit in bulimia nervosa than ordinary supportive counseling.

Dubious therapies. One psychodynamic approach—regrettably still practiced—is “recovered memory therapy.” Therapists who use it claim that childhood sexual abuse or other trauma can cause bulimic symptoms but patients have repressed the memory of these events.²⁰

No methodologically sound evidence has shown that childhood sexual abuse can cause bulimia nervosa years or decades later.²¹ Nor is there acceptable evidence that people can repress the memory of a traumatic experience.²² Therapists administering recovered memory therapy have been subjected to malpractice judgments totaling tens of millions of dollars from suits filed by patients who eventually realized that so-called “recovered” memories were false.²³

Another dubious therapy—eye movement desensitization and reprocessing (EMDR)—also may involve attempts to “recover” memories of putative traumatic events.²⁴ No methodologically sound evidence has shown that EMDR is effective in bulimic patients, and the technique’s theoretical basis is questionable.^24,25

Related resources

• Mental Health Net/Eating disorders. http://eatingdisorders.mentalhelp.net
• National Association of Anorexia Nervosa and Associated Disorders (ANAD). www.altrue.net/site/anadweb/
• Association for Advancement of Behavior Therapy. www.aabt.org

Drug brand names

• Bupropion • Wellbutrin
• Citalopram • Celexa
• Desipramine • Norpramin
• Fluoxetine • Prozac
• Lithium • Lithobid, Eskalith
• Nortriptyline • Pamelor, Aventyl
• Sertraline • Zoloft
• Topiramate • Topamax
• Trazodone • Desyrel
• Liothyronine • Cytomel
• Venlafaxine • Effexor

Disclosure

Dr. Pope receives research support from Ortho-McNeil Pharmaceuticals and is a consultant to Solvay Pharmaceuticals and Auxilium Pharmaceuticals.

Dr. Hudson receives research support from and is a consultant to Eli Lilly & Co. and Ortho-McNeil Pharmaceuticals.

Thousands of scientific papers have been written about bulimia, but not all patients receive effective treatments that produce remission.

To set the record straight and help psychiatrists avoid undertreating bulimia, this article discusses:

• evidence for using antidepressants, even when patients are not “depressed”
• merits of psychotherapies, including those shown to work and those that can harm
• augmentation therapies that can help increase response from partial to full remission.

Initial evaluation

Diagnosis. Bulimia nervosa is characterized by eating binges, followed by purging behaviors such as self-induced vomiting or laxative abuse^1,2 (Table 1). It affects 1% to 3% of adolescent girls and young women and occurs in women 5 to 10 times more often than in men.

Bulimia is often persistent. About one-half of bulimic patients—including those who have been treated—continue to show eating disorder features on long-term follow-up.^3,4

Psychiatric comorbidity. Most bulimic patients report a history of other psychiatric disorders, especially major depressive and bipolar disorders and anxiety disorders such as panic disorder, social phobia, and obsessive-compulsive disorder (OCD).⁵ Because these psychiatric comorbidities may occur before, during, or after bulimia nervosa, one cannot assume that mood or anxiety disorders are a cause or consequence of bulimia. Instead, bulimia nervosa, mood disorders, and anxiety disorders may be different expressions of a shared etiologic abnormality.

Table 1

DSM-IV-TR diagnostic criteria for bulimia nervosa

Evidence supporting this hypothesis comes from studies showing that these disorders:

• respond to several chemically unrelated families of antidepressants^6,7
• frequently co-occur in individual patients^5,7
• frequently co-aggregate in families.^7-9

We have published this evidence^6,7 and proposed that bulimia nervosa may be one form of a larger underlying disorder, which we termed “affective spectrum disorder.”

Antidepressants are often rapidly effective in treating bulimic symptoms,¹⁰ regardless of whether patients exhibit depressive symptoms. Thus, there is no reason to withhold antidepressant therapy simply because a bulimic patient is not depressed. The term “antidepressant” may be a misnomer; these drugs are effective for numerous conditions, of which depression is only one.

Anorexic symptoms. Co-occurring depressive or anxiety disorders in a bulimic patient will not greatly alter treatment. The antidepressants and psychotherapies typically used to treat bulimia are often equally effective for affective disorders. Co-occurring anorexia nervosa, however, is a more serious concern.

Bulimic patients often display a history of anorexia nervosa; in many cases, the patient develops anorexia nervosa as a teenager and then progresses to bulimia nervosa across several years. Her prognosis is much better if her weight normalizes with the shift to bulimia nervosa, than if her weight remains well below normal for her height. It is unclear why medications and psychotherapy are much less effective in bulimic patients with anorexic symptoms than in those with bulimia alone. Watch for further details on anorexia nervosa as this series continues in future issues of.

Medical considerations. Potential medical complications—mostly consequences of vomiting or laxative use—are important to consider when you assess a bulimic patient:¹

• The acid in vomitus may gradually erode tooth enamel, requiring dental consultation.
• Vomitus may inflame salivary gland ducts, though the swelling is usually benign.
• Frequent vomiting may result in hypokalemia and alkalosis, although aggressive medical treatment usually is not needed.

Ask about ipecac use. To induce vomiting, some patients may abuse ipecac syrup, which can cause cardiomyopathy.¹¹

Inpatient or outpatient? Unless the bulimic patient displays severe and medically dangerous anorexic symptoms, she can usually be treated as an outpatient. However, evaluate her carefully for suicidal ideation—which is not uncommon in bulimia nervosa—and consider inpatient treatment if necessary.

Medication vs. psychotherapy

The relative merits of medication versus psychotherapy in treating bulimia nervosa continue to be debated. The Cochrane Database of Systematic Reviews includes meta-analyses of both drug therapy¹² and psychotherapy¹³ for bulimia nervosa. The 2001 drug therapy review found that “the use of a single antidepressant agent was clinically effective,” with no one drug clearly superior to another. Notably, this review was published before recent findings on topiramate.

The corresponding 2002 review of psychotherapy concludes—somewhat more cautiously—that “there is a small body of evidence for the efficacy of cognitive-behavior therapy in bulimia nervosa and similar syndromes, but the quality of trials is very variable and sample sizes are often small.”

In bulimia nervosa and other psychiatric disorders, comparing psychotherapy with drug therapy is hazardous because several factors bias the comparison in favor of psychotherapy. These factors include an expectational effect, a responsibility effect, and differential generalizability of study results.

Expectational effect. Patients in clinical trials are aware that they are receiving psychotherapy and, presumably, that study investigators hope to demonstrate its efficacy. This might account for much of psychotherapy’s apparent effect, as even placebos can produce 30% to 50% improvement in bulimia.¹⁴

Responsibility effect. If a patient fails to improve in a drug study, she will conclude that the drug has failed. But if she fails to improve in a psychotherapy study, she may conclude that she has failed. Because psychological treatments generally require patients to work in therapy, the patient may feel partially responsible for the outcome. Thus, to avoid cognitive dissonance, she may consciously or unconsciously exaggerate her improvement, both in her own mind and when reporting to treaters.

Differential generalizability. Psychological study protocols, such as administering several months of a behavioral treatment, usually mimic clinical practice fairly well. This is not the case with drug study protocols.

No responsible clinician would inflexibly administer a single dosage of a single drug for a fixed period to every bulimic patient and then declare failure for all nonresponders, as is done in study protocols. In practice, the clinician can offer nonresponders augmentation strategies and additional drug trials. Thus, calculations of bulimia response rates in drug studies substantially understate response to drug therapy in clinical practice.

Table 2

How effective are medications in treating bulimia nervosa?

One also might note that psychological study findings have not “sold” well in the clinical practice marketplace. For example, in a recent survey of more than 220 bulimic women treated with psychotherapy, only 6.9% said they received a full course of cognitive behavioral therapy (CBT)¹⁴ —despite two decades of evidence of its efficacy. By contrast, untested, inefficacious, and possibly harmful psychotherapies for bulimia—including recovered-memory therapy—appear to be thriving.

Recommendation. Interpret with caution any head-to-head comparisons of psychological versus drug therapies—especially when clinical practice recommendations are made. Certain psychological therapies provided by specifically-trained individuals likely do help patients with bulimia nervosa. However, biases inherent to the studies may inflate psychological therapies’ efficacy when compared with that of drug therapy.

Therefore, for a psychiatrist who does not specialize in eating disorders to offer exclusively psychological therapy to a bulimic patient—while withholding or postponing drug therapy—may now be a questionable practice.

Choosing drug therapies

Although consensus is lacking on an optimal treatment trial sequence for bulimia nervosa, we suggest a rational approach based on the evidence and our experience (Algorithm).

First-line antidepressants. A selective serotonin reuptake inhibitor (SSRI) trial is usually the first choice (Table 2), and some data suggest that higher-than-usual dosages may be required. For example, in a large multicenter trial of fluoxetine in bulimia nervosa, 60 mg/d was considerably more effective than 20 mg/d for reducing binge eating behavior and vomiting frequency.¹⁵

Based on our observations, however, we believe that noncompliance or irregular compliance may account for this difference in response. Bulimic patients’ impulsive and obsessional behavior may keep them from taking their medications as prescribed. The higher fluoxetine dosage may therefore have been more effective simply because it ensured adequate plasma levels, even when patients missed or forgot multiple doses.

Augmenting agents. A first antidepressant trial rarely leads to complete remission of bulimic symptoms. This is not a serious concern, however, because many other options are available.

Liothyronine. Partial responders to SSRIs often become complete responders when we add a 10-day trial of liothyronine (T3), 25 μg/d. If this fails, we may try augmenting with lithium carbonate, although bulimic patients are often afraid of weight gain or lithium’s other side effects.

Topiramate. A newer augmentation strategy is to add the anticonvulsant topiramate. Used alone, topiramate demonstrated effectiveness for bulimia nervosa in one placebo-controlled, double-blind trial.¹⁶

Adding topiramate to an antidepressant regimen will likely reduce any remaining bulimic symptoms. In addition, topiramate often produces weight loss—a side effect that bulimic patients usually welcome. It remains unclear whether topiramate’s weight-loss effects might pose a hazard in patients with simultaneous bulimic and anorexic symptoms.

Other antidepressants. If the above strategies fail, other antidepressant options include venlafaxine, tricyclics, and monoamine oxidase inhibitors. Bupropion is not recommended in bulimia nervosa; one trial¹⁷ of this agent resulted in a much higher rate of grand mal seizures in bulimic patients than in patients taking bupropion for depression.

In bulimic patients with concomitant bipolar disorder, the anticonvulsants carbamazepine and valproate often reduce affective and bulimic symptoms. By contrast, the anticonvulsant phenytoin—once thought to be useful in bulimia nervosa¹⁰ —offers little benefit for either bulimic or affective symptoms.

Persistence is important when initial medication trials fail. One unblinded study followed 36 bulimic patients 9 to 19 months after they completed a controlled study with trazodone.¹⁸ Of the 26 patients who tried a second or third antidepressant, 17 (65%) achieved remission of bulimia on follow-up. Of the 10 patients who declined a second or third trial, only 1 (10%) attained remission.

Notably, these study results were obtained before the SSRIs and other newer antidepressants or topiramate became available. Cooperative patients using present-day medications might be able to achieve remission rates that exceed 65%.

Algorithm Proposed treatment approach to bulimia nervosa

How effective are psychotherapies in treating bulimia nervosa?

Psychotherapy

Cognitive-behavioral therapy. CBT—given either individually or in groups—is the most effective psychotherapy for bulimia (Table 3).¹⁹ CBT typically involves 3 to 6 months of helping the patient focus on her bulimic behaviors and on specific attitudes—such as unrealistic preoccupations with being “too fat”—that perpetuate the behaviors.

In practice, unfortunately, few bulimic patients are offered CBT, perhaps because few clinicians are trained in the specific approach used for bulimia nervosa.¹⁹ If you are not trained in using CBT for bulimia and do not have access to colleagues who offer this treatment, you may begin with medication plus simple behavioral treatments, such as:

• offering supportive therapy in the office
• referring patients to self-help groups for persons with eating disorders.

If this strategy fails, encourage patients to consider CBT—even if they must travel some distance to obtain it.

Other specialized psychotherapies. Dialectical behavior therapy and interpersonal psychotherapy appear to be effective in bulimia. Again, however, clinicians who lack training in these techniques or access to local experts may be unable to offer them. Psychodynamic therapy does not appear to offer greater benefit in bulimia nervosa than ordinary supportive counseling.

Dubious therapies. One psychodynamic approach—regrettably still practiced—is “recovered memory therapy.” Therapists who use it claim that childhood sexual abuse or other trauma can cause bulimic symptoms but patients have repressed the memory of these events.²⁰

No methodologically sound evidence has shown that childhood sexual abuse can cause bulimia nervosa years or decades later.²¹ Nor is there acceptable evidence that people can repress the memory of a traumatic experience.²² Therapists administering recovered memory therapy have been subjected to malpractice judgments totaling tens of millions of dollars from suits filed by patients who eventually realized that so-called “recovered” memories were false.²³

Another dubious therapy—eye movement desensitization and reprocessing (EMDR)—also may involve attempts to “recover” memories of putative traumatic events.²⁴ No methodologically sound evidence has shown that EMDR is effective in bulimic patients, and the technique’s theoretical basis is questionable.^24,25

Related resources

• Mental Health Net/Eating disorders. http://eatingdisorders.mentalhelp.net
• National Association of Anorexia Nervosa and Associated Disorders (ANAD). www.altrue.net/site/anadweb/
• Association for Advancement of Behavior Therapy. www.aabt.org

Drug brand names

• Bupropion • Wellbutrin
• Citalopram • Celexa
• Desipramine • Norpramin
• Fluoxetine • Prozac
• Lithium • Lithobid, Eskalith
• Nortriptyline • Pamelor, Aventyl
• Sertraline • Zoloft
• Topiramate • Topamax
• Trazodone • Desyrel
• Liothyronine • Cytomel
• Venlafaxine • Effexor

Disclosure

Dr. Pope receives research support from Ortho-McNeil Pharmaceuticals and is a consultant to Solvay Pharmaceuticals and Auxilium Pharmaceuticals.

Dr. Hudson receives research support from and is a consultant to Eli Lilly & Co. and Ortho-McNeil Pharmaceuticals.