Childhood maltreatment and eating disorder pathology: a systematic review and dose-response meta-analysis.

BACKGROUND: Meta-analyses have established a high prevalence of childhood maltreatment (CM) in patients with eating disorders (EDs) relative to the general population. Whether the prevalence of CM in EDs is also high relative to that in other mental disorders has not yet been established through meta-analyses nor to what extent CM affects defining features of EDs, such as number of binge/purge episodes or age at onset. Our aim is to provide meta-analyses on the associations between exposure to CM (i.e. emotional, physical and sexual abuse) on the occurrence of all types of EDs and its defining features. METHOD: Systematic review and meta-analyses. Databases were searched until 4 June 2016. RESULTS: CM prevalence was high in each type of ED (total N = 13 059, prevalence rates 21-59%) relative to healthy (N = 15 092, prevalence rates 1-35%) and psychiatric (N = 7736, prevalence rates 5-46%) control groups. ED patients reporting CM were more likely to be diagnosed with a co-morbid psychiatric disorder [odds ratios (ORs) range 1.41-2.46, p < 0.05] and to be suicidal (OR 2.07, p < 0.001) relative to ED subjects who were not exposed to CM. ED subjects exposed to CM also reported an earlier age at ED onset [effect size (Hedges' g) = -0.32, p < 0.05], to suffer a more severe form of the illness (g = 0.29, p < 0.05), and to binge-purge (g = 0.31, p < 0.001) more often compared to ED patients who did not report any CM. CONCLUSION: CM, regardless of type, is associated with the presence of all types of ED and with severity parameters that characterize these illnesses in a dose dependent manner.

Is season of birth related to disordered eating and personality in women with eating disorders?

We assessed the relation between season of birth and eating disorder symptoms and personality characteristics in a sample of 880 women with eating disorders and 580 controls from two Price Foundation Studies. Eating disorder symptoms were assessed using the Structured Interview of Anorexic and Bulimic Disorders and the Structured Clinical Interview for DSM-IV. Personality traits were assessed using the Temperament and Character Inventory and the Frost Multidimensional Perfectionism Scale. Date of birth was obtained from a sociodemographic questionnaire. No significant differences were observed 1) in season of birth across eating disorder subtypes and controls; nor 2) for any clinical or personality variables and season of birth. We found no evidence of season of birth variation in eating disorders symptoms or personality traits. Contributing to previous conflicting findings, the present results do not support a season of birth hypothesis for eating disorders.

Reduced plasma leptin concentrations in bulimia nervosa.

Leptin is a protein produced by the ob-ob gene which inhibits food intake. Plasma levels have previously been reported to be altered in obesity and anorexia nervosa (AN) but not bulimia nervosa (BN). We measured fasting plasma leptin levels by radioimmunoassay in 53 subjects carefully studied at NIMH, including 37 women meeting DSM-III-R criteria for BN [10 with concurrent AN (body mass index (BMI)=14.1+/-1.4), 27 without AN (BMI=20.4+/-1.6)] and 16 normal control women (NCs) (BMI=21.1+/-2.0). Patients were medication-free and abstinent from bingeing and purging for three to four weeks prior to study. Plasma leptin levels were significantly correlated to BMI (r=0.41, P<0.002), weight (kg, r=0.43, P<0.001), and percent average body weight (%ABW, r=0.45, P<0.001) in the total group. Plasma leptin levels were lower in the BN subjects (3.4+/-2.5 ng/ml) compared to the NCs (6.1+/-2.6 ng/ml, P<0.001, ANCOVA) even after controlling for BMI and weight. There was no significant difference between BN subjects with AN (n=10, 2.6+/-2.6 ng/ml) and those without AN (n=27, 3.8+/-2.4 ng/ml), despite lower BMI in BN with AN. Furthermore, leptin levels were decreased in BN without AN compared with healthy controls, even though BMI was comparable in these two subgroups. Plasma leptin concentrations were negatively correlated with baseline plasma cortisol levels (n=49, r=-0.49, P<0.001) and positively correlated with prolactin responses following L-tryptophan (n=49, r=0.37, P<0.009) and m-chlorophenylpiperazine (n=52, r=0.24, P<0.09). This is the first known report of decreased plasma leptin levels in BN. The decrement in leptin concentration is not related to BMI, body weight, or the presence or absence of BN. HPA axis activation as well as serotonin dysregulation may be related to decreased leptin levels, which may in turn contribute to disinhibited eating in BN. Although current leptin levels were not correlated with self-reported previous binge frequency, the role of leptin in the pathophysiology of BN deserves further study.

Which comes first in the pathogenesis of bulimia nervosa: dieting or bingeing?

OBJECTIVE: Clinical experience has indicated that dieting usually precedes the onset of binge eating in the development of bulimia nervosa (BN). However, data confirming this in nonclinical, representative samples are lacking. METHOD: Using results obtained from the National Women's Study (NWS), we were able to determine the chronological relationship between age of onset of significant dieting (attempting to lose 15 lbs) and onset of bingeing in 85 respondents who met DSM-III-R criteria for BN. These respondents were a subset of over 3,000 female adult U.S. women who completed a random telephone interview (averaging 40 min and including screenings for rape, sexual molestation, aggravated assault, posttraumatic stress disorder [PTSD], and BN). RESULTS: We found that the age of first serious attempt to diet preceded the age of first binge in 46% of cases. There were no significant differences in histories of victimization experiences among the groups. First binge preceded first serious diet in 37% of cases, and these behaviors occurred during the same age in 17% of cases. DISCUSSION: These data confirm that dieting is more likely to precede binge eating, although binge eating precedes significant dieting in a substantial proportion of bulimic respondents.

Comorbidity of psychiatric disorders and posttraumatic stress disorder.

Posttraumatic stress disorder (PTSD) commonly co-occurs with other psychiatric disorders. Data from epidemiologic surveys indicate that the vast majority of individuals with PTSD meet criteria for at least one other psychiatric disorder, and a substantial percentage have 3 or more other psychiatric diagnoses. A number of different hypothetical constructs have been posited to explain this high comorbidity; for example, the self-medication hypothesis has often been applied to understand the relationship between PTSD and substance use disorders. There is a substantial amount of symptom overlap between PTSD and a number of other psychiatric diagnoses, particularly major depressive disorder. It has been suggested that high rates of comorbidity may be simply an epiphenomenon of the diagnostic criteria used. In any case, this high degree of symptom overlap can contribute to diagnostic confusion and, in particular, to the underdiagnosis of PTSD when trauma histories are not specifically obtained. The most common comorbid diagnoses are depressive disorders, substance use disorders, and other anxiety disorders. The comorbidity of PTSD and depressive disorders is of particular interest. Across a number of studies, these are the disorders most likely to co-occur with PTSD. It is also clear that depressive disorder can be a common and independent sequela of exposure to trauma and having a previous depressive disorder is a risk factor for the development of PTSD once exposure to a trauma occurs. The comorbidity of PTSD with substance use disorders is complex because while a substance use disorder may often develop as an attempt to self-medicate the painful symptoms of PTSD, withdrawal states exaggerate these symptoms. Appropriate treatment of PTSD in substance abusers is a controversial issue because of the belief that addressing issues related to the trauma in early recovery can precipitate relapse. In conclusion, comorbidity in PTSD is the rule rather than the exception. This area warrants much further study since comorbid conditions may provide a rationale for the subtyping of individuals with PTSD to optimize treatment outcomes.

Establishing guidelines for pharmacotherapy trials in bulimia nervosa and anorexia nervosa.

OBJECTIVE: This paper addresses the lack of a standard protocol for pharmacotherapy trials for patients with bulimia nervosa (BN) and anorexia nervosa (AN). METHOD: Twenty-two surveys were sent to established researchers in the field of eating disorders to elicit their opinions regarding medication trials, including baseline laboratory tests, the optimal length/frequency of medication management sessions, and the information that should or should not be included in these sessions. RESULTS: Sixteen of 22 researchers completed and returned the survey. Their answers are the basis of the data presented. DISCUSSION: We propose a battery of screening laboratory tests for both conditions. We suggest 30-45-min initial medication management sessions in both AN and BN trials with 15-min follow-ups to be held weekly for AN subjects, and weekly for 2 weeks, then biweekly for 2 weeks, then monthly, for BN subjects. We also recommend that published trials should include explicit details of medication management.

Comorbidity of bulimia nervosa and alcohol use disorders: results from the National Women's Study.

OBJECTIVE: The nature of the relationship between bulimia nervosa (BN) and alcohol abuse/alcohol dependence (AA/AD) and the extent to which women with BN+AA differ from women with BN-AA were examined in a national sample of women (N = 3,006). METHOD: The sample of was generated by multistage geographic sampling and interviews were conducted by telephone. RESULTS: AA was higher in women with BN compared to women without BN or binge eating disorder, only when the influence of major depressive disorder (MDD) and posttraumatic stress disorder (PTSD) was controlled. Prevalence rates of AA and AD were similar in women with BN, MDD, and PTSD. Analyses indicated that the relationship between BN and AA/AD may be indirect and influenced by associations with MDD and PTSD. Women with BN+AA did not differ from those with BN-AA on most variables concerning victimization, family of origin, and disordered eating. DISCUSSION: Evaluation of MDD and PTSD in women presenting for treatment of BN and/or alcohol use disorders (AUDs) is recommended.

The role of weight control as a motivation for cocaine abuse.

Heavy use of cocaine and alcohol in female cocaine abusers with eating disorders has been reported, but the prevalence and motivation for concurrent substance use has not been well investigated. This study of 37 female and 40 male cocaine abusers demonstrated that almost half of the women used cocaine and/or alcohol as a weight control measure, and 13% of the males did the same. Thirteen (72%) of 18 females endorsing weight-related use of cocaine had a current diagnosis of an eating disorder. Only two males (5%) had a past history of an eating disorder. Eleven (85%) of those women with a current eating disorder endorsed using alcohol as an appetite suppressant. These findings support the need to evaluate weight control motivation in cocaine users and to provide specific treatment aimed at addressing the interaction between the eating disorder and the substance abuse problem.

Family characteristics of binge-eating disorder patients.

Binge-eating disorder (BED) has been described and proposed as a new eating disorder diagnosis. Although studies have examined the characteristics of the family members of patients with anorexia nervosa (AN) and bulimia nervosa (BN), little is known about the characteristics of family members of BED patients. The Family Environment Scale (FES) was administered to 88 patients with a DSM-III-R diagnosis of an eating disorder (23 AN, 45 BN, 20 BN + AN), as well as 43 patients with BED as defined by DSM-IV criteria. Statistically significant differences were found among the groups on the cohesion, expressiveness, and active-recreational subscales of the FES by analysis of variance (ANOVA). On the cohesion subscale, significant differences were noted between AN and BED, p < .019, with AN scoring higher than BED. On the expressiveness subscale, significant differences were noted for BED and BN, p < .016, with BN scoring higher than BED. On the active-recreational subscale, there were significant differences for BED versus BN, BED versus AN, and BED versus AN + BN (p < .0001), with BED scoring lower than all other groups. Comparison of BED data to existing normative data yielded significant differences on the cohesion, expressiveness, conflict, independence, intellectual-cultural, active-recreational, and control subscales, with BED patients scoring higher than controls on the conflict and control subscales and lower than controls on all others. These data represent the first study of family characteristics of BED patients utilizing DSM-IV criteria, and provide a beginning understanding of family factors that may be useful in treatment.

Relationship of childhood sexual abuse and eating disorders.

OBJECTIVE: To review the literature that has examined the relationship between childhood sexual abuse and the eating disorders. METHOD: Each of the five authors reviewed all identified empirical studies to be certain that inclusion/exclusion criteria were met. Two teams of raters then independently reviewed each study to determine whether it supported any of a series of six hypotheses that had been tested in this literature. RESULTS: This review indicates that childhood sexual abuse is a nonspecific risk factor for bulimia nervosa, particularly when there is psychiatric comorbidity. There is some indication that childhood sexual abuse is more strongly associated with bulimic disorders than restricting anorexia, but it does not appear to be associated with severity of the disturbance. CONCLUSION: Childhood sexual abuse is a risk factor for bulimia nervosa with significant comorbidity. Further study of the nature of this relationship is warranted.

Cerebrospinal fluid leptin in anorexia nervosa: correlation with nutritional status and potential role in resistance to weight gain.

Studies in rodents have shown that leptin acts in the central nervous system to modulate food intake and energy metabolism. To evaluate the possible role of leptin in the weight loss of anorexia nervosa, this study compared cerebrospinal fluid (CSF) and plasma leptin concentrations in anorexic patients and controls. Subjects included 11 female patients with anorexia nervosa studied at low weight and after treatment, and 15 healthy female controls. Concentrations of leptin in blood and CSF were measured by RIA. Patients with anorexia nervosa, compared to controls, had decreased concentrations of leptin in CSF (98 +/- 26 vs. 160 +/- 58 pg/mL; P < 0.0005) and plasma (1.75 +/- 0.46 vs. 7.01 +/- 3.92 ng/mL; P < 0.005). The CSF to plasma leptin ratio, however, was higher for patients (0.060 +/- 0.023) than for controls (0.025 +/- 0.007; P < 0.0001). At posttreatment testing, although patients had not yet reached normal body weight, CSF and plasma leptin concentrations had increased to normal levels. These results demonstrate the dynamic changes in plasma and CSF leptin during positive energy balance in anorexia nervosa. The results further suggest that normalization of CSF leptin levels before full weight restoration during treatment of anorexic patients could contribute to resistance to weight gain and/or incomplete weight recovery.

The National Women's Study: relationship of victimization and posttraumatic stress disorder to bulimia nervosa.

OBJECTIVE: In recent years there has been increased interest regarding the role of crime victimization in the development and/or maintenance of eating disorders, particularly bulimia nervosa. METHOD: To examine the relationship among assault, bulimia nervosa, and binge eating disorder, a national, representative sample of 3,006 women completed structured telephone interviews. RESULTS: Lifetime prevalence of completed, forcible rape for respondents with bulimia nervosa was 26.6%, as compared with 11.5% for respondents with binge eating disorder and 13.3% for respondents without bulimia nervosa or binge eating disorder. Compared to respondents without bulimia nervosa or binge eating disorder, aggravated assault history was significantly more prevalent in women with bulimia nervosa (26.8%), as was a lifetime history of posttraumatic stress disorder (36.9%). Characteristics of sexual assault experiences were not associated with dysfunctional eating patterns. Specific types of disordered eating such as compensatory behaviors in bulimia nervosa were associated with higher rates of victimization. CONCLUSIONS: In sum, the significantly higher rates of both sexual and aggravated assault among women with bulimia nervosa compared with women without such a diagnosis support the hypothesis that victimization may contribute to the development and/or maintenance of bulimia nervosa.

The phenomenology of psychosis associated with complex partial seizure disorder.

Psychosis commonly occurs as a direct result of complex partial seizure disorder (CPSD). This organic mental disorder is indeed "complex" and is easily and frequently misdiagnosed as a variety of functional disorders, including schizophrenia, schizoaffective disorder, bipolar illness, psychotic depression, and, at best, "atypical psychosis." However, this important clinical syndrome has several clinical features that suggest its presence and which often permit it to be distinguished from other forms of psychosis. Furthermore, this disorder can be successfully treated with limbic anticonvulsants, with or without neuroleptics and/or lithium, but it is generally refractory to neuroleptic medications alone. In this paper, the author reviews the available literature relevant to the clinical phenomenology and treatment of this topic and illustrates the clinical profiles of 10 treatment-refractory patients admitted to a state hospital with previously undiagnosed psychoses secondary to CPSD. This illness needs to be seriously considered in the differential diagnosis of severely ill patients with atypical psychoses refractory to traditional treatments.

Cataplexy and the switch process of multiple personality disorder.

A case history is presented of an 18-year-old male with dissociative disorder and polysubstance abuse. The patient was observed to switch between three personalities, and the personality changes were often associated with symptoms of cataplexy. Both dissociative episodes and cataplexy are associated with strong affective stimuli. Similar reports in the literature are briefly reviewed.

Zolpidem-induced psychosis.

Zolpidem is reported to be a safe and effective hypnotic agent for the short-term treatment of insomnia. There are several case reports of zolpidem causing psychotic reactions in patients with no history of psychosis. We report two additional cases in which zolpidem was implicated in psychotic reactions characterized by auditory and visual hallucinations as well as delusional thinking. Both patient's symptoms resolved with the discontinuation of zolpidem use. It appears that our cases share several features in common with the other reported cases. All were female, there appeared to be some dose dependency involved, and the adverse event resolved fairly quickly upon zolpidem discontinuation. Zolpidem should be used at the lowest effective dose for the least amount of time as necessary. Female patients may possibly require smaller doses. In patients manifesting new-onset or unexplained psychotic symptoms, zolpidem use should be considered in the differential diagnosis.

Studies of serotonin function in anorexia nervosa.

Neuroendocrine, temperature, test-meal, and psychometric responses are reviewed following challenges with the post-synaptic 5-HT receptor agonist m-chlorophenylpiperazine (m-CPP), the 5-HT precursor L-tryptophan (L-TRP), and placebo in 12 patients with anorexia nervosa (AN) and 16 healthy controls. A subset of the AN patients (n = 8) were rechallenged 3-4 weeks after attaining a predetermined goal weight. AN patients had blunted prolactin (PRL) responses to both m-CPP and L-TRP at low-weight and at goal-weight in comparison to controls, although there was a tendency toward normalization with weight gain. There were trends for blunted growth hormone (GH) responses following both L-TRP and m-CPP in the low-weight but not the goal-weight AN patients. Cortisol (CORT) responses following m-CPP and L-TRP were not significantly different among any of the groups. Temperature and test-meal measures were largely unaffected by serotonergic agents in the patients, although m-CPP decreased meal size in the controls. Psychometric responses were variable and are briefly described. Taken together, these findings indicate that responsiveness in post-synaptic hypothalamic-pituitary serotonergic pathways is altered in AN patients. Although there were some trends toward normalization of responsiveness following goal-weight attainment, many differences tended to persist in the patients despite an average increase of 13 kilograms. These may represent changes in serotonergic function at levels in the CNS "above" the hypothalamus.