A controlled study of lifetime prevalence of affective and other psychiatric disorders in bulimic outpatients

The authors used structured diagnostic interviews to assess the lifetime prevalence of psychiatric disorders, by DSM-III criteria, among 70 women: 51 outpatients with active bulimia and 19 nonpatient subjects with remitted bulimia. Comparison groups consisted of 24 female outpatients with major depression and 28 nonpsychiatric control subjects. The active and remitted bulimic subjects closely resembled each other, with high lifetime rates of major affective disorder, anxiety disorders, and substance use disorders. Atypical depression was equally common among subjects with major affective disorder in all groups. These results are consistent with previous studies suggesting a phenomenologic relationship between bulimia and major affective disorder.     

Sleep EEG in bulimia

To evaluate the sleep electroencephalogram (EEG) characteristics of bulimia, all-night sleep EEGs were performed on 11 women meeting DSM-III criteria for bulimia. Comparison groups consisted of young women outpatients with major depression (n = 44) and young normal women (n = 20). The sleep EEGs of the bulimic patients were largely indistinguishable from those of the normal controls, except for a trend toward increased rapid eye movement (REM) density in the first REM period among the bulimic subjects. No differences in any sleep EEG measure were observed between bulimic patients with major depression and those without affective disorder. By contrast, the outpatients with major depression displayed marked sleep continuity disturbances, as well as significantly increased REM intensity and REM density, as compared to normal controls. Implications of these results with respect to the hypothesis that bulimia is related to major affective disorder are discussed.     

Is bulimia associated with borderline personality disorder? A controlled study

To test the hypothesis that bulimia may be associated with borderline personality traits or frank borderline personality disorder, the authors administered the Diagnostic Interview for Borderlines (DIB) under blind conditions to 52 bulimic subjects, 22 outpatient controls with a history of major depression but no history of an eating disorder, and 13 nonpsychiatric controls. Only 1.9% of bulimic subjects met revised DIB criteria for borderline personality disorder, and mean DIB scores were almost identical in bulimic subjects and controls with major depression. These results argue against a relationship between bulimia per se and borderline personality disorder or traits.     

Neuroendocrine responses to m-chlorophenylpiperazine and L-tryptophan in bulimia.

Preclinical and clinical evidence supports a theory of serotonin (5-hydroxytryptamine [5-HT]) dysregulation in bulimia. We therefore studied the prolactin (PRL) and cortisol responses following challenges with the postsynaptic 5-HT receptor agonist m-chlorophenylpiperazine (m-CPP), 0.5 mg/kg orally, the 5-HT precursor L-tryptophan, 100 mg/kg intravenously, and placebo in a group of 28 normal weight bulimic patients and 16 healthy controls. Patients with bulimia, regardless of the presence of major depression, had significantly blunted PRL responses following m-CPP administration compared with those in controls. In contrast, only bulimic patients with concurrent major depression had significantly blunted PRL responses following L-tryptophan administration compared with those in nondepressed bulimic patients and controls. Cortisol responses following m-CPP were not significantly different for bulimic patients vs controls, although there was a trend toward blunted cortisol responses following L-tryptophan administration in the depressed bulimic patients. These differences in neuroendocrine responses were not related to differences in age, percent of average body weight, medications, time of day, peak plasma drug levels, or baseline estradiol levels. Seasonal variations in PRL responses to both agents were identified, although covariation for season did not alter the group differences. The PRL responses following m-CPP administration were inversely correlated to baseline cortisol levels in the bulimic patients, but not in the controls, suggesting a dampening effect by hypothalamic-pituitary-adrenal axis dysfunction on postsynaptic 5-HT receptor sensitivity. The reasons for the differing hormonal responses to these two serotonergic agents may relate to differential involvement of presynaptic and postsynaptic mechanisms, 5-HT receptor subtypes, and anatomical loci of action. The blunted PRL responses to m-CPP administration suggest that postsynaptic 5-HT receptor sensitivity is altered in bulimia nervosa, and that similar alterations in 5-HT receptors at or above the level of the hypothalamus may contribute to binge eating and other behavioral symptoms.     

Postremission predictors of relapse in women with eating disorders

OBJECTIVE: The authors sought to evaluate patterns and predictors of relapse among women with eating disorders. METHOD: Interviews were conducted biannually to annually to assess symptoms of eating disorders, axis I disorders, treatment, and psychosocial function on a weekly basis for women diagnosed with anorexia nervosa (N=136) or bulimia nervosa (N=110) and prospectively followed for 9 years. At the last follow-up, 229 (93%) of the subjects had been retained in the study group. RESULTS: Relapse occurred in 36% of the women with anorexia nervosa and 35% of the women with bulimia nervosa. Women with intake diagnoses of anorexia nervosa, restricting subtype, tended to develop bulimic symptoms during relapse, whereas women with intake diagnoses of anorexia nervosa, binge-purge subtype, or bulimia nervosa tended to return to bulimic patterns during relapse. Greater body image disturbance contributed to a risk of relapse in both eating disorders, and worse psychosocial function increased the risk of relapse in bulimia nervosa. CONCLUSIONS: These results may explain the long-term efficacy of interpersonal therapy for bulimia nervosa and suggest that focused body image work during relapse prevention may enhance long-term recovery from eating disorders.     

Bulimic family dynamics: role of parents' personality--a controlled study with the Temperament and Character Inventory

The numbers of studies on the familial environment and personality of bulimic women have increased in recent years and results have revealed interesting features. In this study, we evaluated the temperament and character traits of patients with bulimia nervosa (BN) and their mothers and fathers, and we analyzed the correlation of temperament and character traits among members of these bulimic families. Finally, we tested the ability of the Temperament and Character Inventory (TCI) to discriminate between normal controls and bulimic subjects, their parents, and their families. Using the TCI, temperament and character features of 28 bulimic patients and their parents (23 fathers and 28 mothers) were analyzed and then compared with a control group of 29 women and their 27 fathers and 29 mothers. Data suggest that both temperament and character factors are involved in BN. Bulimic individuals were high in harm avoidance and low in self-directedness. Their mothers were distinguished by low self-directedness. The fathers were low in persistence. Harm avoidance of bulimic women positively correlated with harm avoidance and negatively with self-directedness of their mothers. The bulimic family had low self-directedness as a common denominator observed in all family members. The observation that both temperament and character have important roles in the etiopathogenesis of bulimia nervosa has important treatment ramifications. The TCI was useful in discriminating between normal controls and bulimic subjects, their parents, and the whole family.     

Effects of an antiandrogenic oral contraceptive on appetite and eating behavior in bulimic women

High androgen levels in women with bulimia nervosa may promote bulimic behavior. The aim of the present study was to investigate the effects of an antiandrogenic oral contraceptive (OC) on appetite and eating behavior in women with bulimia nervosa compared to healthy controls. Twenty-one women with bulimia nervosa and 17 healthy controls matched for age and body mass index participated in the study. Basal and meal-related appetite and secretions of the satiety peptide cholecystokinin (CCK) and the appetite-stimulating peptide ghrelin were studied before and after 3 months of treatment with an antiandrogenic OC (30 microg ethinyl estradiol combined with 3 mg drospirenone). Bulimic behavior was evaluated in relation to changes in hormone levels. Before treatment, bulimic women had higher frequency of menstrual disturbances, acne and hirsutism and higher levels of testosterone but lower meal-related CCK secretion than controls. OC treatment reduced meal-related hunger and gastric distention in bulimics. CCK secretion in response to the meal was unchanged in bulimic women but decreased in the controls. Ghrelin secretion was comparable between groups and did not change in response to OC treatment. The treatment improved bulimic behavior in relation to a decline in testosterone levels in the entire group. Our results support the suggestion that androgens play a role in bulimic behavior. Treatment with an antiandrogenic OC may serve as a new strategy for treatment of bulimia nervosa and particularly in those patients with hyperandrogenic symptoms.     

The validity of the distinction between bulimia with and without anorexia nervosa

The purpose of this study was to examine the relationship between bulimia in women with anorexia nervosa and bulimia of a similar severity in normal-weight women and to determine the support for the distinction between these two groups according to DSM-III criteria. Results indicated that bulimic women with anorexia nervosa (N = 59) and bulimic normal-weight women (N = 59) resembled each other on most variables and were more similar to each other than to women with anorexia nervosa who rigidly restricted food intake (N = 59). The results failed to support the diagnostic distinction between bulimia in anorexic women and bulimia of equal severity in normal-weight women.     

Bulimia in obese individuals. Relationship to normal-weight bulimia

We compared 23 obese subjects meeting DSM-III criteria for bulimia with 47 obese nonbulimic subjects and 47 normal-weight bulimic subjects using structured diagnostic interviews. The obese bulimic subjects were similar to the normal-weight bulimic subjects but different from the nonbulimic obese subjects in exhibiting a high lifetime rate of major affective disorder. However, the obese bulimic subjects were much less likely than the normal-weight bulimic subjects to use self-induced vomiting as a method of purging. These results suggest that obese individuals with bulimic symptoms may constitute a sizable but little-recognized population. Further studies will be required, however, to assess whether the syndrome of bulimia in obesity represents a valid diagnostic entity.     

The effect of bingeing and vomiting on hormonal secretion

Women who are of normal weight and have bulimia nervosa have multiple neuroendocrine disturbances. The reasons for these neuroendocrine abnormalities are not known, but there are reasons to suspect that bingeing and vomiting behavior could be contributory. It is well known that food consumption in healthy volunteers increases plasma insulin, cortisol, and prolactin secretion and suppresses growth hormone secretion, whereas activation of the emetic reflex increases plasma arginine vasopressin (AVP) secretion. The purpose of this study was to investigate the effects of bingeing and vomiting on these hormones. In comparison with healthy control women consuming a large meal, bulimic patients, when bingeing and vomiting, had an exaggerated secretion of either the amount and/or the duration of insulin, cortisol, and prolactin. Vasopressin secretion was not increased during or after bingeing and vomiting, probably because bulimic subjects do not become nauseated. In addition, bulimic patients had significantly reduced baseline plasma prolactin and possibly elevated baseline cortisol compared with controls. In summary, this study supports the presence of neuroendocrine disturbances in bulimia and raises a question as to whether or not excessive and prolonged food consumption (and/or vomiting) are contributory.     

Outcome and clinical course in inpatient bulimic women: a 2- to 9-year follow-up study

BACKGROUND: One previous follow-up study suggested that inpatient bulimic women do quite poorly; after an interval of 2 to 5 years, only 13% were recovered. To examine the course and outcome of a sample of patients with bulimia nervosa that was severe enough to require inpatient hospitalization, the authors conducted the following study. METHOD: Women (N = 52) with DSM-III-R bulimia nervosa were sought 2 to 9 years after hospitalization. Prior to contact, a retrospective chart review was conducted to determine global functioning and admission diagnoses. At follow-up, patients participated in a 4 to 6 hour interview that assessed current and lifetime Axis I disorders (SCID-I), current Axis II disorders (PDE), eating behaviors (EAT, BSQ, EDI, PSR), global functioning (GAF), social adjustment (SAS-SR), and treatment and medical problems experienced since discharge. To assess the significance of differences between the recovered and the currently bulimic women, Yates-corrected chi-square tests and two-tailed t tests were used. RESULTS: Of the 52 women, 46 were interviewed, 1 had died, and 5 could not be located. Of the 46 interviewed women, 39% had fully recovered, 20% had partially recovered, and 41% were currently bulimic. The likelihood of recovery increased with length of time since discharge. While medical problems related to the bulimia were few, treatment with phenelzine was associated with three reports of serious hypertensive episodes, one of which led to death. Global functioning before hospitalization, lifetime DSM-III-R Axis I diagnoses, and current Axis II diagnoses were not associated with outcome. CONCLUSIONS: These findings suggest that even severely ill bulimic patients have a significant chance of achieving full recovery.     

Are anorexic and bulimic patients depressed?

The presence of depression was assessed in 82 female outpatients with anorexia nervosa and bulimia by means of a structured interview. The Research Diagnostic Criteria for major depressive disorder were met by 55.6% of the anorexic patients and 23.6% of the bulimic patients. On the Extracted Hamilton Depression Rating Scale, 40.7% of the patients with anorexia and 23.6% of those with bulimia had scores in the moderately or severely depressed range.     

Other psychiatric disorders among bulimic patients

Clinical experience suggests that bulimia nervosa is a disorder often accompanied by additional psychiatric symptoms. Based on unstructured clinical interviews, 21 additional Axis I and six Axis II diagnoses were assigned to a sample of 30 bulimic women. The Structured Clinic Interview for DSM-III-Patient Version (SCID-P) generated 47 additional Axis I and 78 Axis II diagnoses. There was little agreement across measures in the assessment of mood disturbance. These results not only suggest a greater degree of pathology in bulimic women, but also raise important questions regarding the measurement and conceptualization of DSM-III-R Axis II disorders.     

Anxiety disorders in subjects seeking treatment for eating disorders: a DSM-IV controlled study

Women who were referred with an eating disorder (ED) were compared with a matched normal control group to answer the following questions: What are the frequencies of anxiety disorders in cases of anorexia and bulimia nervosa diagnosed according to DSM-IV criteria? Are anxiety disorders significantly more frequent among women with an eating disorder than among women from the community? We assessed the frequencies of six specific anxiety disorders among 271 women with a current diagnosis of anorexia or bulimia nervosa and 271 controls, using the Mini-International Neuropsychiatric Interview, French DSM-IV version. A lifetime comorbidity with at least one anxiety disorder was found in 71% of both the anorexic and the bulimic subjects, significantly higher than the percentage of controls with an anxiety disorder. The prevalence was significantly higher in the eating disorder groups than in controls for most types of anxiety disorder, and between 41.8 and 53.3% of comorbid cases had an anxiety disorder preceding the onset of the eating disorder. Anxiety disorders are significantly more frequent in subjects with eating disorders than in volunteers from the community, a finding that has important etiological and therapeutic implications.     

Short-term prospective study of recovery in bulimia nervosa

Thirty women who sought treatment at an eating disorders clinic in a general hospital and were diagnosed as having a current episode of bulimia nervosa were followed prospectively to assess the course of eating behaviors and other psychopathology. At intake, all 30 met DSM-III criteria for bulimia nervosa, and 57% met Research Diagnostic Criteria for an affective disorder. After 6 months of followup, 33% had recovered from the bulimic episode, and 24% had recovered from the affective disorder. One subject had recovered from both disorders. In subjects with an affective disorder at intake, the affective disorder had little effect on the time to recovery from bulimia nervosa, and recovery from bulimia nervosa had little effect on the time to recovery from the affective disorder.     

Self-Harm and Suicide Attempts in Individuals with Bulimia Nervosa

Self-harm has been identified as a clinical feature in at least a subgroup of women with bulimia nervosa. We explored how women with bulimia who engage in self-harming behaviors differ from women with bulimia without self-harming behaviors and from women with bulimia who attempt suicide with lethal intent. We compared features of the eating disorder, Axis I and II comorbidity, and Eating Disorder Inventory and Temperament and Character Inventory (TCI) scores across 19 women with bulimia who engaged in self-harming behaviors, 28 women with bulimia who reported suicide attempts with lethal intent, and 105 women with bulimia with no self-harming or suicidal behaviors. Women with bulimia nervosa and self-harm behaviors reported significantly more laxative abuse and drug abuse in general. Bulimic women with suicide attempts had the highest rates of overall comorbidity across the three groups. Individuals with self-harm scored significantly higher on the self-transcendence scale of the TCI. Bulimic women with self-harming behaviors appear to engage in more drug taking behavior--both associated with the eating disorder (laxatives) and in terms of illicit drug use. Overall, the highest rates of Axis I and Axis II comorbidity were associated with individuals with suicide attempts. High scores on self-transcendence may signal a greater sense of dissociation and disconnectedness in bulimic women who self-harm.     

Self-harm and suicide attempts in individuals with bulimia nervosa

Self-harm has been identified as a clinical feature in at least a subgroup of women with bulimia nervosa. We explored how women with bulimia who engage in self-harming behaviors differ from women with bulimia without self-harming behaviors and from women with bulimia who attempt suicide with lethal intent. We compared features of the eating disorder, Axis I and II comorbidity, and Eating Disorder Inventory and Temperament and Character Inventory (TCI) scores across 19 women with bulimia who engaged in self-harming behaviors, 28 women with bulimia who reported suicide attempts with lethal intent, and 105 women with bulimia with no self-harming or suicidal behaviors. Women with bulimia nervosa and self-harm behaviors reported significantly more laxative abuse and drug abuse in general. Bulimic women with suicide attempts had the highest rates of overall comorbidity across the three groups. Individuals with self-harm scored significantly higher on the self-transcendence scale of the TCI. Bulimic women with self-harming behaviors appear to engage in more drug taking behavior--both associated with the eating disorder (laxatives) and in terms of illicit drug use. Overall, the highest rates of Axis I and Axis II comorbidity were associated with individuals with suicide attempts. High scores on self-transcendence may signal a greater sense of dissociation and disconnectedness in bulimic women who self-harm.     

Salivary gland enlargement and elevated serum amylase in bulimia nervosa.

BACKGROUND: Clinical reports have described salivary gland enlargement in bulimia nervosa, particularly in patients with elevated serum amylase concentration. The goal of the current study was to provide a controlled comparison of salivary gland size in patients with bulimia nervosa and healthy volunteers. METHODS: Subjects included 17 women with bulimia nervosa and 21 healthy female control subjects. Dimensions of the parotid and submandibular salivary glands were estimated by ultrasonography. Blood samples for amylase measurement were obtained after overnight fast. RESULTS: Parotid gland size was enlarged 36% in patients with bulimia nervosa in comparison to control subjects (p < .01). For the patient group, salivary gland size was significantly correlated with frequency of bulimic symptoms and with serum amylase concentration. CONCLUSIONS: These results provide new quantitative data demonstrating increased salivary gland size in bulimia nervosa. Further studies are needed to evaluate factors responsible for salivary gland enlargement and hyperamylasemia in this disorder.     

Obsessive-compulsive disorder symptoms before and after recovery from bulimia nervosa.

OBJECTIVE: Obsessive-compulsive disorder (OCD) symptoms are common in people who are ill with bulimia nervosa. However, little is known about whether OCD symptoms persist after long-term recovery from bulimia. METHOD: Thirty-one female patients with bulimia nervosa, 29 women who had been recovered from bulimia for more than 1 year, and 19 healthy female comparison subjects completed the Yale-Brown Obsessive Compulsive Scale, which measures OCD-like symptoms. Items related to symptoms of core eating disorders were omitted from the Yale-Brown scale. RESULTS: The Yale-Brown scale scores of the women with bulimia (mean = 13.1, SD = 10.6) and those who had recovered from bulimia (mean = 7.9, SD = 7.0) were significantly higher than the scores of the normal comparison subjects (mean = 1.9, SD = 2.6). Women with bulimia and those who had recovered from bulimia had similar Yale-Brown scale scores and endorsed similar Yale-Brown scale target symptoms, such as obsessions related to symmetry and exactness. CONCLUSIONS: OCD symptoms persist after recovery from bulimia. Moreover, the types of OCD symptoms experienced by bulimia patients do not vary dramatically with improvement in bulimic symptoms. Persistent OCD symptoms after recovery from bulimia raise the possibility that these behaviors are trait-related and contribute to the pathogenesis of bulimia.     

Sleep of depressed patients with hypersomnia: a 24-h polygraphic study

Sleep of 6 depressed patients with hypersomnia was studied during their depressed phase and their remitted phase using 24-h polygraphic recording. Nine normal subjects were studied as the controls. The latency to sleep onset of the depressed patients was significantly shorter than that of the remitted patients and that of the control subjects. The total sleep time of the depressed patients was significantly longer than that of the remitted patients as well as that of the controls. This increase in sleep time of the depressed patients was mainly due to the increased sleep in day time. The intervals between sleep onset and start of each sleep stage, the relative percentage of individual sleep stages, REM latency and REM density of the depressed patients were not significantly different from those of the remitted patients.