The serotonin hypothesis of bulimia nervosa: theory and evidence

Efficacy of pharmacotherapies for bulimia nervosa: a systematic review and meta-analysis

OBJECTIVE

The main purpose was to evaluate the efficacy and tolerability of different medications used to treat bulimia nervosa (BN).

METHODS

Randomized controlled trials (RCTs) were identified from published sources through searches in PubMed, Cochrane Library, Web of Science, and Embase from inception to November 2022. Primary outcomes were changes in the frequency of binge eating episodes and vomiting episodes from baseline to endpoint. Secondary outcomes were differences in the improvement of scores in depressive symptoms, tolerability (dropout due to adverse events) and weight change.

RESULTS

The literature search ultimately included 11 drugs, 33 studies and 6 types of drugs, 8 trials with TCAs (imipramine, desipramine), 14 with SSRIs (fluoxetine, citalopram and fluvoxamine), 6 with MAOIs (phenelzine, moclobemide and brofaromine), 3 with antiepileptic drugs (topiramate), 1 with mood stabilizers (lithium), and 1 with amphetamine-type appetite suppressant (fenfluramine). The reduction in binge eating episodes was more likely due to these drugs than the placebo, and the SMD was -0.4 (95% CI -0.61 ~ -0.19); the changes in the frequency of vomiting episodes (SMD = -0.16, 95% CI -0.3 ~ -0.03); weight (WMD = -3.05, 95% CI -5.97 ~ -0.13); and depressive symptoms (SMD = -0.32, 95% CI -0.51 ~ -0.13). However, no significant difference was found in dropout due to adverse events (RR = 1.66, 95% CI 1.14 ~ 2.41).

CONCLUSIONS

This meta-analysis indicates that most pharmacotherapies decreased the frequency of binge-eating and vomiting episodes, body weight, and depressive symptoms in BN patients, but the efficacy was not significant. In each drug the efficacy is different, treating different aspects, different symptoms to improve the clinical performance of bulimia nervosa.

Repetitive transcranial magnetic stimulation in bulimia nervosa: preliminary results of a single-centre, randomised, double-blind, sham-controlled trial in female outpatients

BACKGROUND

Bulimia nervosa (BN) is often associated with depressive symptoms and treatment with antidepressants has shown positive effects. A shared deficient serotonergic transmission was postulated for both syndromes. The left dorsolateral prefrontal cortex was argued to regulate eating behaviour and to be dysfunctional in eating disorders.

METHODS

Fourteen women meeting DSM-IV criteria for BN were included in a randomised placebo-controlled double-blind trial. In order to exclude patients highly responsive to placebo, all patients were first submitted to a one-week sham treatment. Randomisation was followed by 3 weeks of active treatment or sham stimulation. As the main outcome criterion we defined the change in binges and purges. Secondary outcome variables were the decrease of the Hamilton Depression Rating Scale (HDRS), the Beck Depression Inventory (BDI) and the Yale-Brown Obsessive Compulsive Scale (YBOCS) over time.

RESULTS

The average number of binges per day declined significantly between baseline and the end of treatment in the two groups. There was no significant difference between sham and active stimulation in terms of purge behaviour, BDI, HDRS and YBOCS over time.

CONCLUSION

These preliminary results indicate that repetitive transcranial magnetic stimulation (rTMS) in the treatment of BN does not exert additional benefit over placebo. A larger number of patients might clarify a further role of rTMS in the treatment of BN.

Eating behavior of bulimics, self-identified binge eaters, and non-eating-disordered individuals: what differentiates these populations?

This article reviews and critiques the eating behavior literature comparing the binge and non-binge-eating episodes of three populations of normal-weight women: bulimics, self-identified binge eaters, and non-eating-disordered women. The specific behaviors evaluated are number of calories consumed during different types of eating episodes, frequency of binge eating, number of eating episodes, rate of food consumption, the macronutrient composition of the food ingested, and context and duration of eating. Differences in these populations' eating behavior are analyzed in terms of their theoretical contribution. It is concluded that differences in the observed behavior of these groups are consistent with restraint theory, purge opportunity, and the forbidden foods hypothesis. Conversely, results do not support carbohydrate craving theory or a deficit in the satiety mechanisms of bulimics. Suggestions for future research are presented.

A randomized controlled trial of fluoxetine and cognitive behavioral therapy for bulimia nervosa: short-term outcome

This study compared and combined fluoxetine and individual cognitive behavioral therapy in the treatment of bulimia nervosa. Participants were 76 women who sought treatment at the Eating Disorders Program of the Toronto Hospital and who met DSM-III-R criteria for bulimia nervosa. Subjects were randomly assigned to receive fluoxetine alone, cognitive behavior therapy alone, or the two in combination and were treated over 16 weeks. Short-term outcome revealed that all three treatment conditions were associated with clinical improvement across a wide range of parameters. The combination of pharmacotherapy and psychotherapy was superior to pharmacotherapy alone on specific parameters and there was no statistically significant advantage to the combination over psychotherapy alone. Limitations to the study include the absence of a placebo pill group and a waiting list control group as well as a substantial dropout rate across all three treatment conditions.

Suicidality in eating disorders: clinical and psychological correlates

Although many authors have described eating disorders as often being associated with suicidal feelings and behavior, few studies to date have evaluated the prevalence and characteristics of suicidal behaviour in eating disordered patients. In the present study, in which a consecutive series of 495 out-patients was studied, 13% of the patients reported at least one suicide attempt and 29% reported current suicidal ideation; 26% of attempters reported multiple attempts. A history of suicide attempt was more prevalent among binge-eating/purging anorexics and among purging bulimics than in the other subgroups. In cases with anorexia nervosa, suicide attempters were older, had a longer illness duration, weighed less, had more often used drugs and/or alcohol and tended to be more obsessive than non-attempters. In cases with bulimia nervosa, attempters presented with more psychiatric symptoms and had more frequently been sexually abused.

Factors that may influence future approaches to the eating disorders

The past thirty years have seen dramatic changes in our approach to the eating disorders. These include marked improvements in the recognition and understanding of the disorders, and the reduction of the morbidity and mortality of anorexia and bulimia nervosa. Bulimia nervosa itself, now the focus of so much attention, was described less than twenty years ago. While these gains, including those in the level of knowledge and appreciation by the public, have been transmitted to the benefit of patients, there remain significant difficulties in treating people with eating disorders. Critical research may address gaps in knowledge and lead to more efficacious treatments. At the same time, however, changes in the health care system are occurring rapidly and these will have an effect on both research and treatment. The purpose of the present article is to highlight scientific and systemic factors that the authors feel will play an important role in how people with eating disorders are responded to in the coming years. In addressing the eating disorders we are confining our remarks to anorexia and bulimia nervosa, and not to obesity; the latter would require consideration of many other issues, beyond the scope of this review.

The hormonal response to intravenous 5-hydroxytryptophan in bulimia nervosa

In recent years evidence has accumulated to implicate a disturbance in serotonin function in the eating disorder bulimia nervosa. This study employs a neuroendocrine technique to assess the effect of intravenous 5-hydroxytryptophan, the immediate precursor to serotonin, on peripheral hormones in bulimia nervosa subjects and controls. Blunted prolactin and growth hormone responses were observed among bulimia nervosa subjects; the possible pathophysiology and implications of the findings are discussed.

Pharmacotherapy of the eating disorders: a commentary

During the last 10 yr, evidence has accumulated which indicates that the eating disorders of bulimia and anorexia nervosa (BN and AN) may be differentially affected by pharmacological treatment. Although the efficacy of drug treatment alone (relative to nonpharmacological approaches) has been debated, there is support for the generalization that all types of antidepressant medications have proven efficacious for bulimia but not for anorexia. These clinical observations are consistent with an extensive body of research concerning the regulation of ingestion, which indicates that the neurotransmitter serotonin plays an important role in mediating satiety. Such considerations have led to the "serotonin-hypothesis of bulimia," which postulates that BN represents an underlying "hyposerotonergic" condition and, conversely, that AN represents a "hyperserotonergic" state. Recently, however, two independent studies have shown that the antidepressant fluoxetine, which selectively blocks the synaptic re-uptake of serotonin, provided significant therapeutic benefit for anorexic patients. The implications of these apparently anomalous results for the "serotonin-hypothesis of BN" are discussed in an attempt to gain insight into the present pharmacotherapy of the eating disorders.

Toward a unified theory of serotonin dysregulation in eating and related disorders

Patients with eating disorders (EDs) exhibit several clinical features and biologic findings indicative of serotonin (5-hydroxytryptamine, 5-HT) dysregulation. These include feeding disturbances, depression and suicide, impulsivity and violence, anxiety and harm avoidance, obsessive-compulsive features, seasonal variation of symptoms, as well as disturbances in neuroendocrine and vascular tissues, as well as other neurochemical systems linked to 5-HT, such as temperature. This review attempts to integrate available results from controlled studies in humans, with particular focus on cerebrospinal fluid (CSF), platelet and plasma studies, as well as pharmacologic challenge strategies using a variety of serotonergic agents. Taken together, these findings support the concept of altered post-synaptic, hypothalamic 5-HT receptor sensitivity in bulimia nervosa (BN), regardless of the presence of anorexia nervosa (AN) or major depression (MD), although these conditions may be associated with other disturbances in 5-HT function, perhaps pre-synaptic ones. The observation that different response measures of 5-HT function in the same subjects may be simultaneously increased, decreased and no different in patients compared to controls is consistent with a 5-HT dysregulation hypothesis. It may be that a variety of psychobiological stressors, such as dieting, binge-eating, purging, drug abuse, photoperiodic changes, as well as psychosocial-interpersonal stressors, perturb and interact with an already vulnerable 5-HT system.

Findings from the Seasonal Pattern Assessment Questionnaire in patients with eating disorders and control subjects: effects of diagnosis and location

We previously reported a high degree of seasonality as measured by the Seasonal Pattern Assessment Questionnaire (SPAQ) in 41 patients with eating disorders (ED) compared with control subjects and patients with five other affective spectrum disorders. To clarify the relationship of the specific ED diagnosis and latitude to seasonal variation in a larger sample, we administered the SPAQ to 159 women with ED as defined by DSM-III-R. Subtype diagnoses were as follows: bulimia nervosa (BN), n = 109; anorexia nervosa (AN), n = 30; BN+AN, n = 20. Patients were studied at three locations: National Institute of Mental Health (NIMH), n = 46; Medical University of South Carolina (MUSC), n = 53; University of Michigan (UM), n = 60. The control group comprised 50 female normal volunteers studied at NIMH. There was a statistically significant difference in Global Seasonality Scale (GSS) scores among the four diagnostic groups, and all ED subtypes had significantly higher GSS scores than control subjects after post hoc Bonferroni t tests. Higher GSS scores were also found in patients with BN+AN compared with patients with BN or AN alone. The patterns of change were similar to those observed in winter seasonal affective disorder (SAD). Thirteen percent of the total sample of ED patients met SPAQ criteria for winter SAD, with 2.5% each for summer SAD and subsyndromal SAD. UM BN patients (latitude 42 degrees N) had higher GSS scores and a higher prevalence of winter SAD than MUSC BN patients (latitude 33 degrees N), but this difference was not statistically significant. These data support the hypothesis that ED and SAD may involve similar pathophysiological mechanisms, possibly related to serotonin dysregulation.

Effects of chronic brofaromine administration on biogenic amines including sulphatoxymelatonin and acid metabolites in patients with bulimia nervosa

Brofaromine, a selective and reversible inhibitor of monoamine oxidase-A (MAO-A) was given to 19 women while 17 received placebo for 8 weeks. All met DSM III-R criteria for bulimia nervosa, a psychiatric disorder in which uncontrolled overeating episodes are accompanied by purging activities and extreme concerns about body shape and weight. The following indices were measured: plasma and urinary phenylacetic acid (PAA), homovanillic acid (HVA), vanillylmandellic acid (VMA); plasma tryptamine (T), beta phenylethylamine (PE), and 5-hydroxyindoleacetic acid (5-HIAA) and urinary 6-sulphatoxymelatonin (aMT6s). PE levels remained the same but T showed a trend toward elevation over time. Twenty-four hour levels of urinary aMT6s in BN patients were higher at week 4 when compared to baseline and week 8. There was a significant reduction in plasma VMA and HVA over time during treatment with brofaromine and both plasma HVA and VMA were significantly lower for the brofaromine group compared to placebo at week 4. Plasma 5-HIAA was significantly higher for the brofaromine group after 8 weeks when compared to placebo. Urinary VMA decreased significantly from baseline to week 4 with a partial elevation at 8 weeks. Urinary VMA was also significantly lower in patients on brofaromine at week 4. This study verifies that brofaromine complies with predicted MAO-A inhibiting patterns in a clinical population.

Is migraine related to the eating disorders?

Migraine and the eating disorders, particularly bulimia nervosa, share some common demographics, phenomenology, psychopathology, and treatments. Bulimics also appear to be more sensitive to the induction of severe migrainous headaches than controls following challenge with the 5-HT agonist, m-chlorophenylpiperazine (m-CPP), but not placebo or L-tryptophan. This supports a common pathophysiological relationship involving postsynaptic 5-HT dysfunction between these disorders. In order to further explore the possible relationship between eating disorders and migraine, we administered a modified version of the Diagnostic Survey of the Eating Disorders (DSED) and the Eating Disorders Inventory (EDI) to a group of female migraine patients attending the Medical University of South Carolina (MUSC) Neurology Clinic (n = 34). Of the 34 migraine patients surveyed, 88% reported dieting behavior, 59% reported binge eating, and 26% reported self-induced vomiting during their lifetimes. Compared to the responses of a group of normal female controls (n = 577), patients with migraine had elevated scores on four of the eight subscales of the EDI: Body Dissatisfaction (p < or = .02), Perfectionism (p < or = .01), Interpersonal Distrust (p < or = .02), and Ineffectiveness (p < or = .06). These findings support the hypothesis that common pathophysiological mechanisms, perhaps involving 5-HT dysregulation, may be involved in these two disorders.

Tridimensional Personality Questionnaire and serotonin in bulimia nervosa

To determine the relationship between Tridimensional Personality Questionnaire (TPQ) scales and bulimia nervosa, TPQ scores of 27 bulimic women, age range 21-59, were compared with values for an age-matched sample of 128 normal control women drawn from the national norming sample by Przybeck. Scores for Novelty Seeking and Harm Avoidance were significantly higher, while scores for Reward Dependence were significantly lower for the bulimic women. A stepwise regression model of severity of purging on TPQ selected Novelty Seeking and a composite depression score, with Novelty Seeking being the stronger of the two predictors. Whole blood serotonin levels did not relate to TPQ scores or to purging frequency.

Impulsivity and dyscontrol in bulimia nervosa: is impulsivity an independent phenomenon or a marker of severity?

Fifty-eight normal-weight DSM-III-R bulimia nervosa patients were compared with 27 normal controls on measures of bulimia nervosa, depression, impulsivity, obsessionality and impulse control (dyscontrol) behaviours. Patients scored higher than controls on all these measures. Almost half the patient cohort met Lacey's multi-impulsive bulimia criteria. When bulimic symptoms were controlled for in an analysis of covariance, multi-impulsive bulimics did not score significantly differently to non-impulsive bulimics on psychometric measures of impulsivity, obsessionality and depression. The concept of impulsivity is critically reviewed and newer concepts linking obsessionality and impulsivity are explored.

A placebo-controlled trial of d-fenfluramine in bulimia nervosa

d-Fenfluramine is a 5-HT agonist which decreases food intake and excessive carbohydrate intake in humans. A placebo-controlled trial of d-fenfluramine (45 mg/day) was conducted in 43 patients with bulimia nervosa. The patients entered an eight-week trial of medication during which they also received cognitive-behavioural therapy. Treatment response was assessed using food diaries to record eating behaviour, and self-rating questionnaires to measure psychopathology. The drug trial, and a follow-up assessment after a further eight weeks, were completed by 39 patients. Abnormal eating behaviour and psychopathology improved significantly in both the d-fenfluramine and placebo groups during the treatment trial. The study failed to show that the addition of d-fenfluramine affords an advantage over brief psychotherapy alone. Although d-fenfluramine is effective in suppressing the overeating, excessive snacking, and excessive carbohydrate consumption which are frequently found in overweight or obese patients, this study suggests that the drug is not an effective treatment for bulimia nervosa.

Is chronic low self-esteem the cause of eating disorders?

Low self-esteem occurs commonly in patients with an eating disorder, a term which includes patients with both anorexia nervosa and bulimia nervosa. In this hypothesis it is proposed that chronic low self-esteem is a necessary prerequisite to the development of an eating disorder, and that chronic low self-esteem is the final common pathway through which the multiple aetiological factors involved in the causation of eating disorders act. Thus, eating disorders can best be viewed as a 'symptom' of chronic low self-esteem. This hypothesis is able to account for the recent increase in the incidence of eating disorders, and the increased incidence of depression in eating disordered patients. Suggestions for research to test this hypothesis are made.

Eating disorders and alcohol abuse in women

Theory and empirical evidence support a relationship between the eating disorders (anorexia nervosa and bulimia nervosa) and alcoholism. This study examines the co-prevalence and characteristics of these disorders among two populations of adult women: those presenting for treatment of alcoholism and those referred to a specialized eating disorders programme. Twenty-two of 73 females (30.1%) with alcohol problems met psychometric cut-off scores for eating disorder, while 25 of the 96 eating disorder females (26.9%) gave psychometric evidence of alcohol dependence. These rates exceed general population norms. While certain clinical and psychometric features distinguish subgroups with both disorders, the basis for co-prevalence and the implications for treatment are unknown.