Chapter 59. Treatment of Eating Disorders

Print: Chapter 59. Treatment of Eating Disorders http://www.psychiatryonline.com/popup.aspx?aID=435228&print=yes…

1 of 9

10/05/2009 16:39

Print Close Window

59. Stewart Agras: Chapter 59. Treatment of Eating Disorders, in The American Psychiatric Publishing Textbook of Psychopharmacology, 4th

Edition. Edited by Alan F. Schatzberg, Charles B. Nemeroff. Copyright ©2009 American Psychiatric Publishing, Inc. DOI:

10.1176/appi.books.9781585623860.435224. Printed 5/10/2009 from www.psychiatryonline.com

Textbook of Psychopharmacology >

Chapter 59. Treatment of Eating Disorders

TREATMENT OF EATING DISORDERS: INTRODUCTION

Eating disorders are seen frequently in the clinic, reflecting a combined prevalence in women for anorexia nervosa

(AN), bulimia nervosa (BN), and binge-eating disorder (BED) of about 3.5% for the full disorder and 6% if

subthreshold disorders are included (Hudson et al. 2007). Males are affected less frequently; about 10% of all cases of

AN and BN are in males, with the proportion rising to about 30% for BED in clinical samples. Because much comorbid

psychopathology is associated with each of these disorders, including current major depression in about 25% of cases,

the treatment plan must take any such disorders into account. Two forms of treatment, psychopharmacological and

psychotherapeutic, are effective in the treatment of BN and BED, and possibly in AN. Hence, determining how to

sequence or combine treatment modalities is an important issue.

HISTORICAL BACKGROUND

Case histories of starvation, binge eating, and purging have been documented, mostly as rare curiosities, for centuries.

The histories of saintly women in the thirteenth to fifteenth centuries who had all the symptoms of AN, including

self-induced vomiting, are particularly well documented both in their own writings and in the accounts of others. For

these women, starvation was in the service of their religious beliefs (Bynum 1987). Cases of BN were rarely seen in

the clinic until their relatively sudden increase throughout the Western world in the late 1970s (Garner et al. 1985).

This apparent increase in the prevalence of BN has been attributed to increasing societal pressures on women to

maintain a thin body shape. Hence, the motivation for excessive dieting may vary according to the cultural pressures

of the times.

Research into the psychopathology and treatment of the eating disorders was relatively slow to develop as compared

with such research in depression and the anxiety disorders. This was probably because of the low prevalence of AN

and the relatively recent increase in the number of cases of BN. Moreover, BED was only recently recognized as a

syndrome, although there is now a large body of literature on the condition. Despite this slow start, sufficient

controlled treatment trials are now available to provide guidance to the clinician. This is particularly true for BN, for

which psychopharmacological and psychotherapeutic studies began at the same time. Because of the similarity

between BN and BED, treatments successful for BN were then applied to BED with some success.

BULIMIA NERVOSA

BN has its onset in late adolescence or early adult life, with a prodromal period characterized by dissatisfaction with

body shape and a fear of becoming overweight, followed by dietary restriction and weight loss. Sooner or later,

periods of dietary restriction are followed by episodes of binge eating experienced as a loss of control over dietary

intake. This, in turn, further aggravates dissatisfaction with body shape and fears of weight gain. Ultimately, the

bulimic patient discovers purging, usually in the form of self-induced vomiting, with or without laxative or diuretic

use, excessive exercise, or (less commonly) fasting; and in rare cases in the form of chewing food and spitting it out.

DSM-IV-TR (American Psychiatric Association 2000) distinguishes two forms of BN: purging and nonpurging types, the

latter characterized by the use of exercise or fasting rather than other compensatory behaviors. The implications of

this classification for treatment are unknown. Medical complications of BN are relatively rare; the most serious are

potassium depletion and dental caries. Other complications include salivary gland enlargement and exercise injuries.

Comorbid psychopathology includes major depression; anxiety disorders, including obsessive-compulsive disorder,

social phobia, and panic disorder; alcoholism; and personality disorders, particularly those in the Cluster B spectrum.

Assessment

Assessment should begin with a history of the development of disordered eating, including the psychosocial factors

involved in its development. Areas that should be explored include binge eating, purging methods, exercise, and

concerns about weight and shape. Eating binges comprise two features: a feeling of loss of control over eating and the

eating of a large amount of food. Loss of control appears to be facilitated by the experience of negative affect often

deriving from faulty interpersonal interactions (Agras and Apple 2007; Agras and Telch 1998). An objective binge

involves eating an amount of food equivalent to an intake of two or more meals. Such binges are required to meet

criteria for the diagnosis. Subjective binges consist of a feeling of loss of control but eating less than the required

amount for an objective binge. These binges are often quite small and may involve eating a “forbidden” food. The most

common method of purging is self-initiated vomiting. It is important to inquire about the use of ipecac to facilitate

vomiting because of its toxic cardiovascular effects. The next most frequent method of purging is the use of laxatives.

Diuretics also may be used, but less frequently than laxatives. Chewing food and spitting it out and the use of enemas

as purgative methods are occasionally seen. Exercise is also used frequently in an attempt to control weight and

shape. It is important to distinguish such exercise from normal exercise patterns. The most common distinguishingPrint: Chapter 59. Treatment of Eating Disorders http://www.psychiatryonline.com/popup.aspx?aID=435228&print=yes…

2 of 9

10/05/2009 16:39

feature is an exercise regimen that results in exhaustion or that is compulsively adhered to and that would cause

anxiety if it were omitted from the daily routine.

In addition to the history, an electrolyte panel is needed, particularly to check potassium levels, which are low and

require correcting in about 5% of individuals with BN. A hematocrit is also useful because anemia may be present.

Serum amylase levels also may be elevated. If the patient is not receiving regular dental care, then a referral for such

care should be considered because of the erosion of dental enamel and periodontal disease that frequently accompany

bulimia. Finally, the assessment should document both past and present comorbid psychiatric disorders because these

conditions may have to be taken into account in planning treatment. For example, current major depression may

interfere with the patient’s ability to adhere to treatment and should be treated together with the eating disorder.

Pharmacological Treatment

Antidepressants

The use of antidepressants in the treatment of BN was sparked by the observation that depression is often a comorbid

feature of the disorder (Pope and Hudson 1982). In 1982, two groups of researchers conducted small-scale

uncontrolled studies indicating that both tricyclic antidepressants and monoamine oxidase inhibitors reduced binge

eating and purging (Pope and Hudson 1982; Walsh et al. 1982). These observations were followed by a series of

double-blind, placebo-controlled studies confirming the utility of antidepressants in treating BN, at least in the short

term.

A wide range of antidepressant drugs have been found effective, including imipramine (Agras et al. 1987; Mitchell et

1. 1990; Pope et al. 1983), desipramine (Agras et al. 1991; Barlow et al. 1988; Blouin et al. 1989; Hughes et al.

1986), phenelzine (Walsh et al. 1988), brofaromine (Kennedy et al. 1993), trazodone (Pope et al. 1989), fluoxetine

(Fluoxetine Bulimia Nervosa Collaborative Study Group 1992), fluvoxamine (Milano et al. 2005), and citalopram

(Leombruni et al. 2006). Fluoxetine is the only medication approved by the U.S. Food and Drug Administration for the

treatment of BN. In these studies, the median rate of decrease in binge eating and purging was 69%, the median

recovery rate was 32%, and the mean dropout rate was 23%. In one study involving 77 BN patients (Walsh et al.

2006b) that examined the rate of decline in bulimic symptoms with desipramine, the authors found that those unlikely

to respond to the antidepressant could be reliably identified after 2 weeks of treatment.

Antidepressants are prescribed for BN at the same dosages used for treating depression, with the exception of

fluoxetine, for which a dosage of 60 mg/day was found to be more effective than 20 mg/day in reducing binge eating

and purging in a placebo-controlled trial involving 387 bulimic women (Fluoxetine Bulimia Nervosa Collaborative

Study Group 1992). One problem with medication given at times other than bedtime is that a significant amount may

be purged through subsequent vomiting. Side effects and reasons for discontinuation of the various medications are

similar to those observed in the treatment of depression. However, a study of bupropion found that a

higher-than-expected proportion of patients developed grand mal seizures (Horne et al. 1988). The authors concluded

that bupropion should not be used for the treatment of BN.

Overall, most antidepressants appear effective for the short-term treatment of BN, with little difference between them

(Bacaltchuk and Hay 2003). An interesting recent study involving 47 patients with BN (Monteleone et al. 2005)

examined the association of the 5-HTTLPR serotonin transporter genotype with antidepressant response, finding that

those with the long form had a tenfold higher likelihood of attaining remission with a selective serotonin reuptake

inhibitor (SSRI). As the authors pointed out, these results, if replicated, could allow therapists to prescribe SSRIs to

those patients with BN who would be most likely to respond to these agents.

Less is known about the long-term effectiveness of medication. Three small-scale uncontrolled studies found that

about one-third of the patients continuing antidepressant medication over periods of 6 months to 2 years relapsed

(Pope et al. 1985; Pyle et al. 1990; Walsh et al. 1991). In a larger-scale examination of this issue, 147 women with BN

who had decreased their vomiting by at least 50% while taking 60 mg of fluoxetine over an 8-week period were

randomly allocated to continue medication or to be switched to placebo (Romano et al. 2002). A survival analysis

found that the group receiving active medication experienced a longer time to relapse (or dropout) than did the

placebo group. However, it should be noted that at the 3-month assessment, 55% of the fluoxetine group and 78% of

the placebo group had either relapsed or dropped out of the study. At the 12-month follow-up, 83% of the fluoxetine

group and 92% of the placebo group had experienced a relapse. The authors suggest that given these results, a

multimodal approach to the treatment of BN, including cognitive-behavioral therapy (CBT), should be considered.

To date, only one study has compared different lengths of antidepressant treatment, in this case with desipramine.

Patients with BN treated for 16 weeks relapsed to pretreatment levels of binge eating when medication was

withdrawn. On the other hand, those treated for 24 weeks maintained remission after withdrawal and at 1-year

follow-up (Agras et al. 1991, 1994a). This study suggests that patients who respond to antidepressant treatment

should be given a minimum trial of 6 months on medication. For the most part, however, controlled studies of

antidepressants are of relatively short duration, as is the assessment of bulimic symptoms. Both of these factors may

somewhat exaggerate the clinical efficacy of these medications.

Other MedicationsPrint: Chapter 59. Treatment of Eating Disorders http://www.psychiatryonline.com/popup.aspx?aID=435228&print=yes…

3 of 9

10/05/2009 16:39

Although considerable evidence from controlled trials indicates that most antidepressants are useful in the treatment

of BN, few controlled studies of other pharmacological agents have appeared in the literature. However, topiramate

(an anticonvulsant drug) has been evaluated in two controlled trials. In the first of these studies, patients meeting

criteria for DSM-IV-TR bulimia nervosa were allocated at random to treatment with either topiramate ( n = 35) or

placebo (n = 34) over a 10-week period (Hoopes et al. 2003). Twenty-two (63%) of those in the topiramate group

completed the trial, and topiramate was statistically superior in reducing binge eating and purging, and 22% of

completers were recovered at the end of treatment. In the second study (Nickel et al. 2005), 30 patients with BN were

randomly allocated to either topiramate or placebo. Topiramate was statistically superior to placebo in reducing binge

eating and purging; however, no data on remission or recovery were reported. Although no direct comparison between

topiramate and an antidepressant has been made, the data so far suggest that topiramate is about as effective as the

antidepressants in the treatment of BN.

Combined Treatment

CBT for BN was developed in parallel with the use of antidepressants. CBT in either individual or group format has

been shown to be more effective in reducing binge eating and/or purging than placebo (Mitchell et al. 1990),

supportive psychotherapy plus self-monitoring of eating behavior (Agras et al. 1989), stress management (Laessle et

1. 1991), behavior therapy (Fairburn et al. 1993), and psychodynamic forms of psychotherapy (Garner et al. 1993;

Walsh et al. 1997). The existence of two different and effective treatments, antidepressant medications and CBT,

naturally led to the question of whether the combined treatments would be more effective than either treatment

alone.

The first study of this question used a randomized 2 x 2 design with four experimental groups: 1) imipramine

combined with group psychosocial treatment, 2) imipramine with no psychosocial treatment, 3) placebo combined

with group psychosocial treatment, and 4) placebo with no psychosocial treatment (Mitchell et al. 1990). Treatment

was preceded by a single-blind placebo washout phase. One hundred and seventy-one women with BN entered

treatment, which lasted for 10 weeks. The psychosocial treatment was an intensive group variant of CBT, with 5 daily

sessions in the first week and 22 treatment sessions overall. The mean daily dosage of imipramine was 217 mg for the

psychosocial treatment group and 266 mg for the group receiving medication alone. As might be expected, the dropout

rate was significantly higher for those receiving medication (34%) compared with those taking placebo (15%).

Imipramine was superior to placebo; however, CBT, with a remission rate of 51%, was superior to imipramine, with a

remission rate of 16%, and combining the two treatments did not result in any additional advantage in reducing binge

eating and purging. The combined treatment was, however, significantly superior to CBT in reducing depression.

In the second study (Agras et al. 1991, 1994a), 71 participants were randomly allocated to one of three groups: 1)

desipramine (mean daily dosage = 168 mg), 2) CBT, and 3) combined treatment. Half of the desipramine participants

were withdrawn from medication at 16 weeks and the remainder at 24 weeks. CBT lasted for 24 weeks. Eighteen

percent of the participants stopped taking desipramine before the end of treatment, compared with a treatment

dropout of 4.3% of those participants receiving CBT. CBT, with a 48% remission rate, was significantly superior to

desipramine, with a 33% remission rate, in reducing the frequency of binge eating and purging, and the combined

treatment was no more effective than CBT alone. At 1-year follow-up of 61 of the original 71 patients, 77% of the

combined treatment group were abstinent, compared with 54% of those receiving CBT alone (Agras et al. 1994a). This

difference was not statistically significant. However, receiving desipramine alone for 24 weeks was the most

cost-effective approach in terms of the cost of treatment per recovered patient at 1-year follow-up (Koran et al. 1995).

Another study involving 120 women with BN used a more sophisticated medication regimen consisting of desipramine

followed by fluoxetine if the first medication was either ineffective or poorly tolerated (Walsh et al. 1997). It is

important to note that the two-medication combination was used by two-thirds of the patients assigned to active

medication, suggesting that a two-medication combination is closer to clinical reality than the use of a single

medication. This study used a five-cell design: CBT combined with placebo or active medication, psychodynamically

oriented therapy combined with placebo or active medication, and medication alone. CBT (plus placebo) was more

effective than psychodynamic therapy (plus placebo) in reducing both binge eating and purging. The average dosage

of desipramine was 188 mg/day and of fluoxetine was 55 mg/day. Forty-three percent of the patients receiving

medication dropped out of the study, compared with 32% of those receiving psychotherapy. Patients receiving active

medication (in combination with psychological treatments) reduced binge eating significantly more than did those

receiving placebo. Finally, antidepressant medication combined with CBT was superior to medication alone in reducing

purging frequency. Of those receiving CBT plus medication, 50% were in remission, compared with 25% of those

receiving medication alone. These findings suggest that the combination of CBT plus antidepressant medication may

be the most effective approach to the treatment of BN. A meta-analysis confirmed this impression (Bacaltchuk et al.

2000).

Treatment After Psychotherapy Failure

In a small-scale double-blind, randomized, controlled trial, participants who had failed to respond to either CBT or

interpersonal therapy in a multisite trial were randomly allocated to either fluoxetine 60 mg/day or placebo (Walsh et

1. 2000). Twenty-three participants entered the study with a median of 22 binges and 30 purges over a 4-week

period. Of those receiving fluoxetine, 38% were abstinent (over a 4-week period) at the end of treatment compared

with none in the placebo group. This finding suggests that fluoxetine may be useful for those who do not respond toPrint: Chapter 59. Treatment of Eating Disorders http://www.psychiatryonline.com/popup.aspx?aID=435228&print=yes…

4 of 9

10/05/2009 16:39

psychological treatments.

Comprehensive Treatment

Patients with BN are best treated as outpatients, unless there are either medical or psychiatric reasons for

hospitalization (e.g., an intercurrent physical illness or a comorbid psychiatric disorder requiring hospitalization, such

as major depression with suicidality). One reason that outpatient treatment is useful for the BN patient is that gains

made in the hospital may not carry over to the patient’s natural environment, where more complex food stimuli and

greater stress are present than in the hospital.

The research evidence to date suggests that the combination of antidepressant medication and CBT is likely to be

somewhat more effective than either therapy alone. Because CBT is more effective than antidepressant medication,

having fewer dropouts than medication, in the ideal case, this should be the first therapy offered to the patient.

However, CBT is not always available, and in such circumstances, medication will be the only choice. In addition,

patient preferences for one or another treatment should be taken into account.

The flow chart in Figure 59–1 presents an algorithm as guidance to the overall treatment of BN. The first decision to

make is whether the patient has current major depression, which is seen in approximately 25% of bulimic patients

presenting for treatment. Because depressive symptoms can interfere with the conduct of CBT for BN, antidepressant

medication should precede CBT in such patients. When the patient has sufficiently recovered from depression, the

eating disorder should be reevaluated. If the patient has not recovered from the eating disorder, then CBT should be

added.

FIGURE 59–1. Flow chart depicting different treatment sequences for bulimia nervosa (BN).Print: Chapter 59. Treatment of Eating Disorders http://www.psychiatryonline.com/popup.aspx?aID=435228&print=yes…

5 of 9

10/05/2009 16:39

CBT = cognitive-behavioral therapy.

As shown in Figure 59–1, after 4 weeks (six sessions) of therapy, if the reduction in purging with CBT is less than

70%, an antidepressant should be added. This algorithm is based on the findings of a multisite study involving 194

women with BN, which found that the initial treatment response to CBT predicted outcome reasonably well (Agras et

1. 2000). Those reducing purging less than 70% after 4 weeks of treatment were more likely to be nonresponders. If

there is an inadequate response to antidepressant treatment or a relapse, an alternative antidepressant should be

used. For those who complete CBT with insufficient improvement, despite having reduced their purging at session 6,

then an antidepressant should be advised.

BINGE-EATING DISORDER

Although the association between binge eating and obesity has been noted from time to time in case reports in the

literature, it was not until the upsurge of research into the psychopathology and treatment of BN that systematic

attention was paid to BED. The principal features of BED are episodes of binge eating at a frequency of at least 2 days

a week for 6 months, marked distress caused by binge eating, and binge eating that does not occur during the course

of BN or AN. Purging does not occur in this condition, although about 10% of the patients with BED have a history of

1.  

Between 1% and 2% of women in the general population meet criteria for BED (Bruce and Agras 1992). In clinical

populations, the ratio of women to men with BED is approximately 3:2, the highest rate for men for any eating

disorder. Although obesity is not a requirement for the diagnosis of BED, there is a substantial overlap between BED

and obesity. Studies have shown that about one-quarter of obese subjects have symptoms that meet criteria for BED

and that the prevalence of binge eating increases as body mass index increases (Marcus et al. 1985; Spitzer et al.

1993; Telch et al. 1988). Because binge eating often precedes the onset of becoming overweight, binge eating may be

a risk factor for obesity and the multiple health problems associated with being overweight. Moreover, the syndrome

is associated with comorbid psychopathology similar to that of BN and causes much distress; hence, it is an entity

deserving of treatment in its own right. One study that compared individuals with BED with weight-matched

non-binge-eating obese individuals found that subjects with BED were significantly more likely to receive diagnoses of

major depression (51%), panic disorder (9%), and borderline personality disorder (9%) than were those without BED

(Yanovski et al. 1992).

Pharmacological Treatment

Antidepressants

Double-blind, placebo-controlled studies suggest that antidepressants are at least as useful in the treatment of BED as

in BN. Early placebo-controlled studies found desipramine to be effective in reducing binge eating, with an abstinence

rate of 60% (McCann and Agras 1990). Studies of SSRIs suggest moderate efficacy for fluoxetine, sertraline, and

citalopram (Appolinario and McElroy 2004; McElroy et al. 2000). Other studies have found no effect for fluoxetine on

binge eating (Devlin et al. 2005; Grilo et al. 2005b). These conflicting results may be due to the relatively high placebo

response observed in some studies. There has also been considerable variability in weight losses experienced by

patients with BED treated with antidepressants, from essentially no weight loss to several pounds. Although the

McCann and Agras (1990) study found that antidepressants reduced binge eating, patients who stopped binge eating

did not lose weight. In another controlled study, 108 overweight women with BED received 3 months of CBT, followed

by 6 months of weight-loss treatment combined with desipramine. No additive effect of desipramine on binge eating

was found; however, women in the medication group lost significantly more weight (4.8 kg) than those in the

comparison group (Agras et al. 1994b).

The serotonin–norepinephrine reuptake inhibitor sibutramine (15 mg/day) and the selective norepinephrine reuptake

inhibitor reboxetine (8 mg/day) also appeared to be useful in the treatment of BED in open-label studies (Shapira et

1. 2000; Silveira et al. 2005). Weight losses were relatively large with both of these medications. A placebo-controlled

study of sibutramine in 60 obese women with BED confirmed these early studies (Appolinario et al. 2003), with 52%

achieving abstinence from binge eating compared with 32% in the placebo group. Weight losses were 7.4 kg for the

sibutramine group compared with a small increase in weight for the placebo group.

Other Medications

Anticonvulsants such as topiramate and zonisamide also appear useful in the treatment of BED (McElroy et al. 2006).

A multisite study in which more than 400 participants with BED were allocated at random to either topiramate or

placebo provided further evidence of the efficacy of topiramate (McElroy et al. 2007). The median dosage of

topiramate was 300 mg/day. Dropouts were equivalent between groups (29% topiramate; 30% placebo). Fifty-eight

percent of those in the topiramate group and 29% in the placebo group were in remission at the end of the study

period. The mean weight loss was 4.5 kg in the topiramate group versus a small gain in the placebo group. The most

common side effects specific to topiramate were paresthesia and difficulty concentrating. Hence, topiramate leads to a

reasonable rate of remission combined with substantial weight loss.

Finally, a placebo-controlled study comprising 50 overweight participants with BED compared orlistat (120 mg three

times daily) with placebo, both groups receiving an abbreviated form of CBT (Grilo et al. 2005a). At the end ofPrint: Chapter 59. Treatment of Eating Disorders http://www.psychiatryonline.com/popup.aspx?aID=435228&print=yes…

6 of 9

10/05/2009 16:39

treatment, 64% of those in the orlistat group and 36% of those in the placebo group were in remission. The

proportions achieving at least a 5% weight loss were 36% for orlistat and 8% for placebo. However, after

discontinuation of both treatments, there was no difference in abstinence rates between groups (52% in both groups).

Comprehensive Treatment

BED presents three problems to the clinician: binge eating, overweight, and comorbid psychopathology, particularly

depression. Hence, a comprehensive treatment should address all of these problems. There are few direct comparisons

of psychotherapy and medication, and the situation is further complicated by the larger placebo responses found in

BED as compared with BN, probably accounting for the lack of efficacy of pharmacological agents in some studies. For

patients who prefer to try medication, it appears on present evidence that sertraline and topiramate are the most

effective in terms of reducing both binge eating and weight. Less is known about medications that may add to the

effects of psychotherapies such as CBT and interpersonal therapy, both of which are associated with substantial

reductions in binge eating that appear to be well maintained, although weight losses are small. Hence, it appears

reasonable to augment psychotherapy with either sertraline or another SSRI, topiramate, or orlistat.

ANOREXIA NERVOSA

AN is a relatively rare disorder characterized by marked weight loss (at least 15% below ideal body weight), intense

fear of gaining weight, disturbance in the experience of body shape (i.e., feeling fat in the face of marked weight

loss), and (in females) amenorrhea. It is the most lethal psychiatric disorder. Follow-up studies document an

aggregate mortality rate of about 5.6% per decade; about half of these deaths are a result of suicide, and the

remainder are largely due to cardiovascular instability (Casiero and Fishman 2006). Because of the chronicity of the

condition, it has become apparent that identification and treatment of the disorder early in its course are essential. A

specific family therapy for adolescents that aims to help parents take charge of their child’s eating appears to be

successful in both the short and the long term, with about 70% of adolescent patients with anorexia recovered both at

the end of treatment and at follow-up (Lock et al. 2005, 2006).

Most patients with anorexia can be treated as outpatients. However, treatment can be difficult because of the patient’s

reluctance to gain weight. Weight should be monitored at every outpatient visit, and it is important that weight be

measured in a hospital gown to prevent the use of lead weights to which some patients with anorexia resort. Other

methods of inflating weight are less easy to detect, such as drinking large quantities of water before being weighed.

Indications for hospitalization include weight less than 75% of ideal body weight for age and height, heart rate below

40 beats/minute, blood pressure below 90/60 mm Hg, potassium levels below 3 mEq/L, temperature below 97°F, and

very rapid weight loss. In addition, because of the associated psychopathology in this disorder, the usual indications

for hospitalization for severe psychopathology should be followed.

Because the disorder is rare, it is difficult in any one center to acquire an adequate sample size for a study in a

reasonable time; thus, satisfactory randomized, double-blind medication trials are difficult to accomplish. Moreover,

medication trials should be long enough and use optimal medication dosages to adequately show effects in this

chronic relapsing disorder. Unfortunately, few trials meet these criteria; many are of short duration, use inadequate

dosages of medication, or have been carried out in inpatients.

Pharmacological Treatment

Most studies of antipsychotic agents in the treatment of AN, including chlorpromazine, pimozide, and sulpiride,

showed no evidence of efficacy (Dally and Sargant 1960; Vandereycken 1984; Vandereycken and Pierloot 1982).

Recent case reports, however, suggest that risperidone may lead to weight gains (Newman-Toker 2000). The effects

of olanzapine were studied in 34 patients with anorexia nervosa receiving day care treatment who were allocated to

active drug or placebo. The mean dosage of olanzapine was 6.61 mg/day. Those allocated to olanzapine recovered

more quickly than those on placebo and gained about 0.6 kg more than those on placebo. Obsessive thinking was also

decreased in those receiving olanzapine, but there were no other differences in psychopathology between groups

(Bissada et al. 2008). Further studies with larger sample sizes are needed. Moreover, the clinical impression is that

olanzapine is not well accepted by patients with AN.

An important study found that fluoxetine was not effective in hospitalized patients with AN (Attia et al. 1998). In this

study, 31 women hospitalized with AN participated in a 7-week randomized, double-blind trial of fluoxetine at a mean

dosage of 56 mg/day. Four patients in each group terminated the trial early. Although all patients in the study showed

improvement, no significant differences were seen between active medication and placebo. In addition, there was no

apparent effect of medication on depression or obsessional symptoms. This study suggested that fluoxetine had no

effect over and above that of an inpatient program and adds to the consistent failure to show a beneficial effect of

antidepressant medication during the period of weight regain.

Despite the findings of an earlier small-scale outpatient study (Kaye et al. 2001), a more recent study of 93 adult

outpatients found no benefit of fluoxetine in either promoting weight maintenance or prolonging time to relapse in a

double-blind, placebo-controlled trial (Walsh et al. 2006a). As is usual in this population, there was a large proportion

of dropouts or early terminations from treatment (51% of fluoxetine-treated and 63% of placebo-treated patients). A

fairly high proportion of patients were dissatisfied with treatment. The very high dropout rates make a statistical

comparison between groups difficult because of the large amount of data being carried forward in an intent-to-treat

analysis. Nonetheless, the only difference between groups was a statistical advantage for fluoxetine in reducingPrint: Chapter 59. Treatment of Eating Disorders http://www.psychiatryonline.com/popup.aspx?aID=435228&print=yes…

7 of 9

10/05/2009 16:39

anxiety levels.

Given the findings that fluoxetine confers no benefit for adult patients with AN either during the weight gain period in

hospital or during outpatient treatment, one must conclude that the use of fluoxetine is not indicated in the treatment

of AN except for the treatment of comorbid psychopathology. At this point, however, there have been no satisfactory

studies of SSRIs in adolescents with AN, and such studies would appear warranted given the high priority for

treatment early in the course of AN.

CONCLUSION

The place of psychopharmacological agents in the treatment of BN has been well worked out. Treatment with

sequential trials of different antidepressants should result in abstinence rates of about 40%. The addition of CBT

enhances the effectiveness of antidepressants. It is becoming clear that agents such as topiramate and similar

anticonvulsants are useful in the treatment of BED, with the added advantage of facilitating substantial weight loss in

the overweight patient. In the case of AN, there is little evidence that pharmacological agents are helpful in either

inpatient or outpatient treatment of the adult patient, except to treat comorbid psychiatric disorders. There is

insufficient information regarding adolescent anorexia to provide guidance regarding the use of medication at this

point.

REFERENCES

Agras WS, Apple RF: Overcoming Eating Disorders: Therapist Guide. New York, Oxford University Press, 2007

Agras WS, Telch CF: The effects of caloric deprivation and negative affect on binge eating in obese binge-eating

disordered women. Behav Ther 29:491–503, 1998

Agras WS, Dorian B, Kirkley BG, et al: Imipramine in the treatment of bulimia: a double-blind controlled study. Int J

Eat Disord 6:29–38, 1987

Agras WS, Schneider JA, Arnow B, et al: Cognitive-behavioral and response prevention treatment for bulimia nervosa.

J Consult Clin Psychol 57:215–221, 1989 [PubMed]

Agras WS, Rossiter EM, Arnow B, et al: Pharmacologic and cognitive-behavioral treatment for bulimia nervosa: a

controlled comparison. Am J Psychiatry 159:325–333, 1991

Agras WS, Rossiter EM, Arnow B, et al: One-year follow-up of psychosocial and pharmacologic treatments for bulimia

nervosa. J Clin Psychiatry 55:179–183, 1994a

Agras WS, Telch CF, Arnow B, et al: Weight loss, cognitive-behavioral, and desipramine treatments in binge-eating

disorder: an additive design. Behav Ther 25:225–238, 1994b

Agras WS, Crow SJ, Halmi KA, et al: Outcome predictors for the cognitive-behavioral treatment of bulimia nervosa:

data from a multisite study. Am J Psychiatry 157:1302–1308, 2000 [Full Text] [PubMed]

American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision.

Washington, DC, American Psychiatric Association, 2000

Appolinario JC, McElroy SL: Pharmacologic approaches to the treatment of binge eating disorder. Curr Drug Targets

5:301–307, 2004 [PubMed]

Appolinario JC, Bacaltchuk J, Sichieri R, et al: A randomized double-blind, placebo-controlled study of sibutramine in

the treatment of binge eating disorder. Arch Gen Psychiatry 60:1109–1116, 2003 [PubMed]

Attia E, Haiman C, Walsh BT, et al: Does fluoxetine augment the inpatient treatment of anorexia nervosa? Am J

Psychiatry 155:548–551, 1998 [Full Text] [PubMed]

Bacaltchuk J, Hay P: Antidepressants versus placebo for people with bulimia nervosa. Cochrane Database Syst Rev

(4):CD003391, 2003

Bacaltchuk J, Trefiglio IR, Oliveria P, et al: Combination of antidepressants and psychological treatments for bulimia

nervosa: a systematic review. Acta Psychiatr Scand 101:256–267, 2000 [PubMed]

Barlow J, Blouin J, Blouin A, et al: Treatment of bulimia with desipramine: a double-blind crossover study. Can J

Psychiatry 33:129–133, 1988 [PubMed]

Bissada H, Tasca GA, Barber AM, et al: Olanzapine in the treatment of low body weight and obsessive thinking in

women with anorexia nervosa: a randomized, double-blind, placebo-controlled trial. Am J Psychiatry 165:1281–1288,

2008 [Full Text] [PubMed]

Blouin J, Blouin A, Perez E: Bulimia: independence of antibulimic and antidepressant properties of desipramine. Can J

Psychiatry 34:24–29, 1989 [PubMed]

Bruce B, Agras WS: Binge-eating in females: a population-based investigation. Int J Eat Disord 12:365–373, 1992

Bynum CW: Holy Feast and Holy Fast: The Religious Significance of Food to Medieval Women. Berkeley and Los

Angeles, University of California Press, 1987Print: Chapter 59. Treatment of Eating Disorders http://www.psychiatryonline.com/popup.aspx?aID=435228&print=yes…

8 of 9

10/05/2009 16:39

Casiero D, Fishman WH: Cardiovascular complications of eating disorders. Cardiol Rev 14:227–231, 2006 [PubMed]

Dally PJ, Sargant W: A new treatment of anorexia nervosa. BMJ 1:1770–1773, 1960 [PubMed]

Devlin MJ, Golfein, JA, Petkova E, et al: Cognitive-behavioral therapy and fluoxetine as adjuncts to group behavioral

therapy for binge eating disorder. Obes Res 13:1077–1088, 2005.

Fairburn CG, Jones R, Peveler RC, et al: Psychotherapy and bulimia nervosa: longer-term effects of interpersonal

psychotherapy, behavior therapy, and cognitive behavior therapy. Arch Gen Psychiatry 50:419–428, 1993 [PubMed]

Fluoxetine Bulimia Nervosa Collaborative Study Group: Fluoxetine in the treatment of bulimia nervosa. Multicenter,

placebo-controlled, double-blind trial. Arch Gen Psychiatry 49:139–147, 1992

Garner DM, Olmsted MP, Garfinkel PE: Similarities among bulimic groups selected by weight and weight history. J

Psychiatr Res 19:129–134, 1985 [PubMed]

Garner DM, Rockert I, Davis R, et al: Comparison between cognitive-behavioral and supportive-expressive therapy for

bulimia nervosa. Am J Psychiatry 150:37–46, 1993 [Full Text] [PubMed]

Grilo CM, Masheb RM, Salant SL: Cognitive behavioral therapy, guided self-help and orlistat for the treatment of binge

eating disorder: a randomized double-blind placebo-controlled trial. Biol Psychiatry 57:1193–1201, 2005a

Grilo CM, Masheb RM, Wilson GT: Efficacy of cognitive behavioral therapy and fluoxetine for the treatment of binge

eating disorder: a randomized double-blind placebo-controlled comparison. Biol Psychiatry 57:301–309, 2005b

Hoopes SP, Reimherr FW, Hedges DW, et al: Treatment of bulimia nervosa with topiramate in a randomized,

double-blind, placebo-controlled trial, part 1: improvement in binge and purge measures. J Clin Psychiatry

64:1335–1341, 2003 [PubMed]

Horne RL, Ferguson JM, Pope HG, et al: Treatment of bulimia with bupropion: a multicenter controlled trial. J Clin

Psychiatry 49:262–266, 1988 [PubMed]

Hudson JI, Hiripi E, Pope HG, et al: The prevalence and correlates of eating disorders in the national comorbidity

survey replication. Biol Psychiatry 61:348–358, 2007 [PubMed]

Hughes PL, Wells LA, Cunningham CI, et al: Treating bulimia with desipramine. A double-blind, placebo-controlled

study. Arch Gen Psychiatry 43:182–187, 1986 [PubMed]

Kaye WH, Nagata T, Weltzin TE, et al: Double-blind placebo-controlled administration of fluoxetine in restricting- and

restricting-purging-type anorexia nervosa. Biol Psychiatry 49:644–652, 2001 [PubMed]

Kennedy SH, Goldbloom DS, Ralevski E, et al: Is there a role for selective monoamine oxidase inhibitor therapy in

bulimia nervosa? A placebo-controlled trial. J Clin Psychopharmacol 13:415–422, 1993 [PubMed]

Koran LM, Agras WS, Rossiter E, et al: Comparing the cost-effectiveness of psychiatric treatments: bulimia nervosa.

Psychiatry Res 58:13–21, 1995 [PubMed]

Laessle PJ, Beumont PJV, Butow P, et al: A comparison of nutritional management with stress management in the

treatment of bulimia nervosa. Br J Psychiatry 159:250–261, 1991 [PubMed]

Leombruni P, Amianto F, Delsedime N, et al: Citalopram versus fluoxetine for the treatment of patients with bulimia

nervosa: a single-blind randomized controlled trial. Adv Ther 23:481–494, 2006 [PubMed]

Lock J, Agras WS, Bryson S, et al: A comparison of short- and long-term family therapy for adolescent anorexia

nervosa. J Am Acad Child Adolesc Psychiatry 44:632–639, 2005 [PubMed]

Lock J, Couturier J, Agras WS: Comparison of long term outcomes in adolescents with anorexia nervosa treated with

family therapy. J Am Acad Child Adolesc Psychiatry, 45:666–672, 2006 [PubMed]

Marcus MD, Wing RR, Lamparski DM: Binge-eating and dietary restraint in obese patients. Addict Behav 10:163–168,

1985 [PubMed]

McCann UD, Agras WS: Successful treatment of compulsive binge-eating with desipramine: a double-blind

placebo-controlled study. Am J Psychiatry 147:1509–1513, 1990 [PubMed]

McElroy SL, Casuto LS, Nelson EB, et al: Placebo-controlled trial of sertraline in the treatment of binge eating disorder.

Am J Psychiatry 157:1004–1006, 2000 [Full Text] [PubMed]

McElroy SL, Kotwal R, Guerdjikova A, et al: Zonisamide in the treatment of binge eating disorder with obesity: a

randomized controlled trial. J Clin Psychiatry 67:1897–1906, 2006 [PubMed]

McElroy SL, Hudson JI, Capece JA, et al: Topiramate for the treatment of binge eating disorder associated with obesity:

a placebo-controlled study. Biol Psychiatry 61:1039–1048, 2007 [PubMed]

Milano W, Siano C, Putrella C, et al: Treatment of bulimia nervosa with fluvoxamine: a randomized controlled trial. Adv

Ther 22:278–283, 2005 [PubMed]

Mitchell JE, Pyle RL, Eckert ED, et al: A comparison study of antidepressants and structured intensive groupPrint: Chapter 59. Treatment of Eating Disorders http://www.psychiatryonline.com/popup.aspx?aID=435228&print=yes…

9 of 9

10/05/2009 16:39

psychotherapy in the treatment of bulimia nervosa. Arch Gen Psychiatry 47:149–160, 1990 [PubMed]

Monteleone P, Santonastaso P, Tortorella A, et al: Serotonin transporter polymorphism and potential response to

SSRIs in bulimia. Mol Psychiatry 10:716–718, 2005 [PubMed]

Newman-Toker J: Risperidone in anorexia nervosa. J Am Acad Child Adolesc Psychiatry 39:941–942, 2000 [PubMed]

Nickel C, Tritt K, Muehlbacher M, et al: Topiramate treatment in bulimia nervosa patients: a randomized, double-blind,

placebo-controlled trial. Int J Eat Disord 38:295–300, 2005 [PubMed]

Pope HG, Hudson JI: Treatment of bulimia with antidepressants. Psychopharmacology (Berl) 78:176–179, 1982

[PubMed]

Pope HG, Hudson JI, Jonas JM, et al: Bulimia treated with imipramine: a placebo-controlled double-blind study. Am J

Psychiatry 140:554–558, 1983 [PubMed]

Pope HG, Hudson JI, Jonas JM, et al: Antidepressant treatment of bulimia: a two-year follow-up study. J Clin

Psychopharmacol 5:320–327, 1985 [PubMed]

Pope HG, Keck PE, McElroy SL, et al: A placebo-controlled study of trazodone in bulimia nervosa. J Clin

Psychopharmacol 9:254–259, 1989 [PubMed]

Pyle RL, Mitchell JE, Eckert ED, et al: Maintenance treatment and 6-month outcome for bulimic patients who respond

to initial treatment. Am J Psychiatry 147:871–875, 1990 [PubMed]

Romano SJ, Halmi KA, Sarka NP, et al: A placebo-controlled study of fluoxetine in continued treatment of bulimia

nervosa after successful acute fluoxetine treatment. Am J Psychiatry 159:96–102, 2002 [Full Text] [PubMed]

Shapira NA, Goldsmith TD, McElroy SL: Treatment of binge-eating disorder with topiramate: a clinical case series. J

Clin Psychiatry 61:368–372, 2000 [PubMed]

Silveira RO, Zanatto V, Appolinario JC, et al: An open trial of reboxetine in obese patients with binge eating disorder.

Eat Weight Disord 10:e93–e96, 2005

Spitzer RL, Yanovski S, Wadden T, et al: Binge eating disorder: its further validation in a multisite study. Int J Eat

Disord 13:137–153, 1993 [PubMed]

Telch CF, Agras WS, Rossiter EM: Binge-eating increases with increasing adiposity. Int J Eat Disord 7:115–119, 1988

Vandereycken W: Neuroleptics in the short-term treatment of anorexia nervosa: a double-blind placebo-controlled

study with sulpiride. Br J Psychiatry 144:288–292, 1984 [PubMed]

Vandereycken W, Pierloot R: Pimozide combined with behavior therapy in the short-term treatment of anorexia

nervosa: a double-blind placebo-controlled cross over study. Acta Psychiatr Scand 66:445–450, 1982 [PubMed]

Walsh BT, Stewart JW, Wright L, et al: Treatment of bulimia with monoamine oxidase inhibitors. Am J Psychiatry

139:1629–1630, 1982 [PubMed]

Walsh BT, Gladis M, Roose SP, et al: Phenelzine vs placebo in 50 patients with bulimia. Arch Gen Psychiatry

45:471–475, 1988 [PubMed]

Walsh BT, Hadigan CM, Devlin MJ, et al: Long-term outcome of antidepressant treatment for bulimia nervosa. Am J

Psychiatry 148:1206–1212, 1991 [PubMed]

Walsh BT, Wilson GT, Loeb KL, et al: Medication and psychotherapy in the treatment of bulimia nervosa. Am J

Psychiatry 154:523–531, 1997 [Full Text] [PubMed]

Walsh BT, Agras WS, Devlin MJ, et al: Fluoxetine for bulimia nervosa following poor response to psychotherapy. Am J

Psychiatry 157:1332–1334, 2000 [Full Text] [PubMed]

Walsh BT, Kaplan AS, Attia E, et al: Fluoxetine after weight restoration in anorexia nervosa: a randomized controlled

trial. JAMA 295:2605–2612, 2006a

Walsh BT, Sysko R, Parides MK: Early response to desipramine among women with bulimia nervosa. Int J Eat Disord

39:72–75, 2006b

Yanovski SZ, Nelson JE, Dubbert BK, et al: Association of binge-eating disorder and psychiatric comorbidity in the

obese. Am J Psychiatry 150:1472–1479, 1992

Copyright © 2009 American Psychiatric Publishing, Inc. All Rights Reserved.