Compliance and outcome in anorexia nervosa.

Anorexia nervosa is notoriously difficult to treat, but little is known regarding the relationship of compliance to treatment outcome. We investigated in 41 adolescents who fulfilled DSM-III-R criteria for anorexia nervosa, the relationship between the completion of a standard psychosocial treatment program, subtypes of anorexia nervosa, and outcome as determined by standardized measurements. These adolescents were observed for an average of 32.4 months. Overall, 29 patients (70%) improved considerably, but 10 (24%) were symptomatic, and 2 (5%) remained in poor condition. There were no deaths. Of the 41 patients, 14 (34%) completed our entire treatment program, 15 (37%) received major treatment and failed in the outpatient follow-up phase only, 7 (17%) dropped out of inpatient treatment before its completion, and 5 (12%) refused treatment in our system altogether. Of all the dropouts, 10 received no further treatment. One patient was admitted to hospital elsewhere but again dropped out in the outpatient phase of that program. Seven patients (17%) received further outpatient treatment only, and 9 (22%) received inpatient and outpatient care and seemingly completed their treatment. Treatment completion significantly affected the measures of global clinical functioning and specific psychopathologic features, but only for those patients who completed the initial program. Bulimic patients did considerably worse on follow-up and were less likely to complete treatment. Patients with restricted anorexia nervosa were more likely to complete treatment than those with a bulimic subtype (P = .03). Differential compliance rates in the two subtypes confound the effects of treatment completion and need to be controlled for in future studies. Depression was not associated with noncompliance but, if present, was associated with poor outcome on follow-up and abated in only a third of those in whom it was initially present.     

Plasma transferrin levels in abnormal endocrine states. II. The changes in various endocrine states

The clinical significance of the measurement of plasma transferrin (Tf) in patients with hypophysial disorders was reported in our previous paper. In the present study, we determined plasma Tf levels in 55 patients with various endocrine states and considered their clinical significance compared with plasma somatomedin-C (SM-C) levels. Plasma Tf levels decreased significantly in patients with anorexia nervosa (P less than 0.02), hyperthyroidism (P less than 0.05), primary hypothyroidism (P less than 0.05), and Cushing's syndrome (P less than 0.05), while they were elevated significantly in pregnancy (P less than 0.01) or females using estrogens (P less than 0.05). The former two declines were considered a reflection of the malnutritional state since a significant negative correlation was observed between plasma Tf levels and the percentile deficit from the ideal body weight in patients with anorexia nervosa (P less than 0.01), or between plasma Tf and elevated T3 levels which induce hypermetabolism in patients with hyperthyroidism (P less than 0.01). A significant correlation was observed between the SM-C and Tf levels in these subjects (including normal controls and patients with hypophysial disorders) as a whole (r = 0.79, P less than 0.001). These data indicate that plasma Tf is changeable according to the endocrine and nutritional conditions with good correlation to the SM-C, and it is suggested that Tf also operates as a growth factor in vivo.     

Human pancreatic polypeptide responsiveness to insulin-induced hypoglycemia in anorexia nervosa

Patients with anorexia nervosa occasionally suffer from hypoglycemic comas. We investigated the role of human pancreatic polypeptide (HPP) in insulin-induced hypoglycemia (0.1 U/kg of regular insulin). Ten female patients with anorexia nervosa (20.7 +/- 2.0 years, mean +/- SEM; 34.9 +/- 1.7 kg, mean +/- SEM) and 8 age-matched female controls (20.9 +/- 0.6 years, 51.5 +/- 0.8 kg) were tested. In the patients with anorexia nervosa, testing was performed before and after the restoration of body weight (45.0 +/- 0.8 kg). There was no significant difference in glucose nadir between patients with anorexia nervosa and the control subjects. However, glucose recovery from nadir was delayed in patients with anorexia nervosa. In anorexia nervosa patients, the plasma pancreatic glucagon responses to insulin-induced hypoglycemia did not differ from those of the controls. Results also showed, however, that HPP responses to insulin-induced hypoglycemia were significantly higher in patients with anorexia nervosa than in controls (p less than 0.01). The increased HPP responses were still present after the restoration of body weight in anorexia nervosa patients. A complete body weight recovery or a longer period of time may be required to normalize the HPP response to insulin-induced hypoglycemia in patients with anorexia nervosa, after the restoration of body weight.     

Angiotensin-converting enzyme and anorexia nervosa

Angiotensin-converting enzyme (ACE) activity was measured in 10 patients with anorexia nervosa, 6 with hyperthyroid Graves' disease, and 7 with primary hypothyroidism. Patients with anorexia nervosa had a low serum ACE activity (9.8 +/- 2.2 IU/l), as compared to findings in normal subjects (13.4 +/- 3.5 IU/l) (P less than 0.05). Patients with hyperthyroid Graves' disease had high serum ACE activity (23.7 +/- 5.8 IU/l), as compared to levels in normal subjects (P less than 0.01), and patients with primary hypothyroidism tended to have low serum ACE activity (10.1 +/- 1.8 IU/l), compared to the normal subjects (P less than 0.1). Following weight gain (before; 71.3 +/- 10.2% of ideal body weight, after; 88.7 +/- 5.6% of ideal body weight), serum ACE activity in patients with anorexia nervosa reverted to within the normal range (13.8 +/- 3.5 IU/l), and serum T3 concentration was restored to the normal range (before; 0.7 +/- 0.2 ng/ml, after; 1.1 +/- 0.3 ng/ml). In these patients, ACE activity correlated with the per cent of ideal body weight (P less than 0.05). These data suggest that, in underweight subjects with anorexia nervosa, decreased serum ACE activities may relate to emaciation.     

Serum leptin levels in patients with anorexia nervosa before and after partial refeeding, relationships to serum lipids and biochemical nutritional parameters.

Leptin is a protein hormone produced by adipocytes that provide information about the body fat content. It was previously reported that serum leptin levels were decreased in patients with anorexia nervosa in comparison with healthy control subjects. The aim of our study was to compare serum leptin levels in patients with anorexia nervosa (n=11, initial mean BMI=15.4 kg/m2) before and after partial recovery with control age-matched subjects (n=11, mean BMI= 20.3 kg/m2) and to study the relationships of leptin levels, serum lipids and biochemical nutritional parameters. We found that serum leptin concentrations in patients with anorexia nervosa were significantly reduced in comparison with control subjects (3.61 vs 9.37 ng.ml(-1), p<0.01). Serum cholesterol, triglycerides, total protein and albumin in patients with anorexia nervosa either before or after partial recovery did not differ from the control group. After partial recovery, a significant increase in serum leptin was observed (4.83 vs 3.61 ng.ml(-1), p<0.05), but the values still remained significantly lower than in the control group (p<0.01) Leptin levels correlated positively with the body mass index in the control group and anorexia nervosa group before recovery. The correlation with BMI in the anorexia nervosa group after refeeding was not significant. No significant correlation was found between leptin concentrations and serum lipids, total protein, albumin and prealbumin, respectively. Serum leptin thus represents a sensitive parameter that reflects the nutritional status in patients with anorexia nervosa suitable for long-term follow up during refeeding therapy.     

Human atrial natriuretic polypeptide in plasma of patients with anorexia nervosa

To examine the effects of chronic dehydration and starvation on plasma levels of human atrial natriuretic polypeptide (hANP) in human subjects, the basal level and saline-induced rise of plasma hANP in 7 patients with anorexia nervosa were compared with those in age-matched healthy subjects. The unstimulated level of plasma hANP was markedly high in the patients with anorexia nervosa (patients vs. control; 55.4 +/- 9.0 pg/ml vs. 11.4 +/- 6.1 pg/ml, P less than 0.01). However, no significant increase of plasma hANP in the anorectic patients was observed in response to saline-infusion, while a 3-fold increase over the basal level of plasma hANP was noted in the saline-infused normal young subjects. These results show that hANP may be secreted to an inadequate extent, hence the release would be resistant to volume-loading. The pathophysiological meaning of such a high plasma concentrations of hANP in anorexia nervosa is the subject of ongoing studies.     

Resting tachycardia, a warning sign in anorexia nervosa: case report

Background

Among psychiatric disorders, anorexia nervosa has the highest mortality rate. During an exacerbation of this illness, patients frequently present with nonspecific symptoms. Upon hospitalization, anorexia nervosa patients are often markedly bradycardic, which may be an adaptive response to progressive weight loss and negative energy balance. When anorexia nervosa patients manifest tachycardia, even heart rates in the 80–90 bpm range, a supervening acute illness should be suspected.

Case presentation

A 52-year old woman with longstanding anorexia nervosa was hospitalized due to progressive leg pain, weakness, and fatigue accompanied by marked weight loss. On physical examination she was cachectic but in no apparent distress. She had fine lanugo-type hair over her face and arms with an erythematous rash noted on her palms and left lower extremity. Her blood pressure was 96/50 mm Hg and resting heart rate was 106 bpm though she appeared euvolemic. Laboratory tests revealed anemia, mild leukocytosis, and hypoalbuminemia. She was initially treated with enteral feedings for an exacerbation of anorexia nervosa, but increasing leukocytosis without fever and worsening left leg pain prompted the diagnosis of an indolent left lower extremity cellulitis. With antibiotic therapy her heart rate decreased to 45 bpm despite minimal restoration of body weight.

Conclusions

Bradycardia is a characteristic feature of anorexia nervosa particularly with significant weight loss. When anorexia nervosa patients present with nonspecific symptoms, resting tachycardia should prompt a search for potentially life-threatening conditions.     

Assessment of the relationship between serum thyroid hormone levels and peripheral metabolism in patients with anorexia nervosa

It has been observed that basal and/or TRH-stimulated serum TSH levels occasionally conflict with the actual values of circulating thyroid hormones in patients with anorexia nervosa. In the present study sixteen female patients with anorexia nervosa during self-induced starvation displayed clinical findings suggesting hypothyroidism, e.g., cold intolerance, constipation, bradycardia, hypothermia and hypercholesterolemia in association with decreased serum total T3 (62.8 +/- 5.2 ng/dl) and T4 (6.6 +/- 0.3 micrograms/dl). Markedly decreased T3 correlated positively with average heart rate (r = 0.5655, P less than 0.025) and negatively with total cholesterol (r = -0.7413, P less than 0.005). This result may suggest that peripheral metabolic state of the underweight anorexics depends considerably upon the serum T3 concentration. Despite decreased total thyroid hormones, free T4 assayed by radioimmunoassay was normal in all five cases examined (1.4 +/- 0.2 ng/dl) and the free T4 index in fifteen cases was normal except in one case. Basal TSH was not increased and TSH response to exogenous TRH was not exaggerated in any. These results may be compatible with a theory that free T4 has a dominant influence on pituitary TSH secretion. Furthermore, glucocorticoids may also have some influence on depressed TSH response, because an inverse correlation between increased plasma cortisol and the sum of net TSH increase after TRH was observed in twelve cases examined. In conclusion, it is suggested that normal sensitivity of peripheral tissues and pituitary thyrotroph to different circulating thyroid hormones is maintained in anorexia nervosa patients even during severe self-induced starvation, and that the metabolic state in these patients is considerably under the influence of circulating T3.     

Abnormal perception of food size in anorexia nervosa

The hypothesis that patients with anorexia nervosa exaggerate the perceived size of food was tested. Video recordings of five items of food and four neutral objects of a similar size were made such that the size of each object increased steadily from half to twice its normal size. Each of the nine objects was placed on a ledge inside a dummy television screen next to the video screen, the food items alternating with the neutral objects, and 20 female patients with anorexia nervosa and 20 female controls matched for age were asked to adjust the size of the video recording to that of the real object. Although there was no overall difference in perceptual accuracy between patients and controls, both groups perceived the food items as being bigger than the neutral objects, the patients exaggerating the size of the food significantly more than the controls.These results imply that patients with anorexia might start eating more easily when admitted if their food were presented as small portions on large plates.     

Classification of feeding and eating disorders: review of evidence and proposals for ICD-11

Current classification of eating disorders is failing to classify most clinical presentations; ignores continuities between child, adolescent and adult manifestations; and requires frequent changes of diagnosis to accommodate the natural course of these disorders. The classification is divorced from clinical practice, and investigators of clinical trials have felt compelled to introduce unsystematic modifications. Classification of feeding and eating disorders in ICD-11 requires substantial changes to remediate the shortcomings. We review evidence on the developmental and cross-cultural differences and continuities, course and distinctive features of feeding and eating disorders. We make the following recommendations: a) feeding and eating disorders should be merged into a single grouping with categories applicable across age groups; b) the category of anorexia nervosa should be broadened through dropping the requirement for amenorrhoea, extending the weight criterion to any significant underweight, and extending the cognitive criterion to include developmentally and culturally relevant presentations; c) a severity qualifier "with dangerously low body weight" should distinguish the severe cases of anorexia nervosa that carry the riskiest prognosis; d) bulimia nervosa should be extended to include subjective binge eating; e) binge eating disorder should be included as a specific category defined by subjective or objective binge eating in the absence of regular compensatory behaviour; f) combined eating disorder should classify subjects who sequentially or concurrently fulfil criteria for both anorexia and bulimia nervosa; g) avoidant/restrictive food intake disorder should classify restricted food intake in children or adults that is not accompanied by body weight and shape related psychopathology; h) a uniform minimum duration criterion of four weeks should apply.     

Effect of implanting bull calves with testosterone propionate, dihydrotestosterone propionate or oestradiol-17 beta prepubertally on the pituitary-testicular axis and on postpubertal social and sexual behaviour.

Twelve non-implanted crossbred bull calves served as controls and 30 crossbred bull calves (10/treatment) were implanted for 82 days, beginning at 34 days of age, to determine the influence of testosterone propionate (TP), dihydrotestosterone propionate (DHTP) and oestradiol-17 beta (E2) on prepubertal and pubertal pituitary-testicular function and on postpubertal social and sexual behaviour. Compared with control bulls, concentrations of serum luteinizing hormone (LH), follicle-stimulating hormone (FSH) and inhibin concentrations were suppressed (P less than 0.01) in all implanted bulls. Testosterone (T) concentration increased (P less than 0.001) in TP-implanted, but decreased (P less than 0.01) in DHTP and E2 bulls during the implant period. LH response to gonadotrophin-releasing hormone (GnRH) challenge during the implant period (2.5 months of age) was less (P less than 0.01) in TP, E2 and DHTP bulls than in controls. A small but significant T response to GnRH occurred in control bulls at 2.5 months of age. LH and T responses to GnRH challenge at 7 months of age (100 days after implant removal) was similar (P greater than 0.20) in control and implanted bulls. Steroid implants administered prepubertally had no effect (P greater than 0.10) on postpubertal social and sexual behaviours, including number of flehmen responses, abortive mounts, services and competitive order score. Body weight did not differ (P greater than 0.10) between treatment groups, but testis size was reduced (P less than 0.01) during the implant period and up to 10 months of age in treated bulls compared with controls. Testes remained smaller in E2-treated bulls up to the end of the study (23 months of age), but daily sperm production and epididymal weight did not differ (P greater than 0.10) between treatment groups at slaughter. Control bulls reached puberty earlier (P less than 0.01; 270 +/- 11 days of age) than did TP (302 +/- 11 days), DHTP (309 +/- 11 days) or E2 (327 +/- 11 days) bulls. Although puberty was delayed in all implant groups, there was no difference in scrotal circumference at puberty (average 28.4 +/- 0.4 cm) between treatment groups. Our findings indicate that TP, DHTP and E2 implants administered prepubertally result in acute suppression of serum LH, FSH and inhibin during the implant period and in post-implant suppression of testis size and delayed puberty in bulls. The lack of treatment effect on behaviour suggests that steroidal programming of sexual behaviour occurs before 1 month of age in bulls.     

The outcome of treatment for anorexia nervosa inpatients who required urgent hospitalization

Background

This study was done to determine which psychosocial factors are related to the urgent hospitalization of anorexia nervosa patients (AN) due to extremely poor physical condition and to evaluate their outcome after inpatient treatment.

Methods

133 hospitalized AN patients were classified into an urgent hospitalization (n?=?24) or a planned hospitalization (n?=?109) group. Multiple regression analysis was done of clinical features, body mass index (BMI), psychological tests [The Minnesota Multiphasic Personality Inventory (MMPI), alexithymia, relationship with parents, and the Eating Disorder Inventory (EDI)]. The effectiveness of treatment was prospectively determined two years after discharge by the Global Clinical Score (GCS). The hospitalized weight gain and the frequency of outpatient visits were evaluated.

Results

Of the factors assessed, only BMI at admission was related to the necessity of urgent hospitalization (β?=?? 1.063, P?=?0.00). The urgent group had significantly more weight loss after discharge and poorer social adaptation on the GCS, even when the patient had a sufficient increase in body weight during inpatient treatment and an equivalent number of outpatient consultations.

Conclusion

None of the parameters of the psychosocial tests studied were significantly different between the groups. The outcome of the urgent group was poor. Two years after discharge they had difficulty maintaining weight and continued to have poor social adaptation.

     

Steroids and neuroendocrine function in anorexia nervosa

Anorexia nervosa is a primarily psychiatric syndrome of self-induced weight loss due to an intense fear of becoming obese. Numerous endocrine abnormalities occur in anorexia nervosa patients, and in many respects these alterations reflects the endocrinology of reduced energy intake. However, the basic mechanisms of those alterations are far from being understood. In an attempt to understand the disrupted mechanisms of the hypogonadotropic hypogonadism of the anorectic state, we studied 10 anorectic women in the acute phase of their illness; all met the DSM III criteria. On each patient, two tests were performed with either saline as control or infusion of the opioid antagonist naloxone, and both LH and FSH levels were measured. Four mg of naloxone as bolus was used, followed by a naloxone infusion of 2 mg/h for 4 h. Compared with the pattern of normal women, naloxone did not increase in the anorectic patients either LH or FSH levels nor pulsatility. This result suggests that endogenous opioid peptides are not implicated in the low gonadotropic situation of anorexia nervosa. An alternative explanation could be that the low estrogenic "milieu" of these patients could mask the opioid action. To test this second possibility, another group of 7 anorectic women after partial weight recovery were challenged with estrogen administration. Compared with the pattern of normal women volunteers, all the anorectic patients but one presented an abnormal response in both LH and FSH levels after estrogen administration. In fact, the negative feedback and the delayed positive feedback of LH after estrogen were absent in these patients. Interestingly enough, the only patient with near-normal LH response to estrogen was considered fully recovered by the Psychiatric Unit. Several alterations in the hypothalamic-pituitary-adrenal axis has been reported in anorexia nervosa. Seven anorectic patients and 7 aged-matched women were challenged by ACTH 1-24, 250 micrograms (i.v.) and the ratio of increments in adrenal steroid products to precursors monitored. ACTH-induced increments in cortisol with respect to increments in 17-OH-progesterone was similar in anorectics and controls. On the contrary, the ratio of increments of androstenedione with respect to increments in 17-OH-progesterone were greater in anorexia nervosa than controls. These results suggest that in anorexia nervosa the 11-beta-21-alpha-hydroxylase system is normal but a deficient 17-20 desmolase system is present. Finally, the altered pattern of GH secretion in anorexia was studied using GHRH (1 microgram/kg) as stimulus of pituitary GH secretion.(ABSTRACT TRUNCATED AT 400 WORDS)     

Interrelationship between serum muscle enzymes and a low T3 in anorexia nervosa

To determine the interrelationship between muscle dysfunction and a low T3 state, both seen in anorexia nervosa, we studied the relationship between the degree of muscle involvement, as assessed by the circulating concentration of the three muscle indicators (CPK, GOT and LDH), and serum T3 in thirty-three patients when they were admitted to the hospital. We also studied the malnutritional state, as assessed by their body weight or serum GH, serum potassium and the degree of hyperactivity exhibited. Additionally, another twelve patients were studied in order to explore the mounding phenomenon which is typically elicited in hypothyroidism. The logarithms of serum CPK and GOT correlated only with the serum T3 concentration (r = -0.35, p less than 0.05; r = -0.41, p less than 0.05; respectively). The logarithm of serum LDH highly correlated with serum T3, the percentage of ideal body weight, and the logarithm of serum GH (r = -0.55, p less than 0.01; r = -0.66, p less than 0.001; r = 0.43, p less than 0.05; respectively). The mounding phenomenon was elicited in ten out of twelve patients. In conclusion, it was implied that a low T3 state was associated with an increase in serum muscle indicators and thus with muscle dysfunction encountered in anorexia nervosa.     

Clinically apparent eating disorders in young diabetic women: associations with painful neuropathy and other complications.

Of 208 young women with insulin dependent diabetes, 15 (7%) had a clinically apparent eating disorder (anorexia nervosa or bulimia), a much higher prevalence than reported in non-diabetic women. Most, but not all, of these patients had a long history of poor glycaemic control. In contrast with previous suggestions, control did not deteriorate after the onset of the eating disorder. There was a high incidence and an early onset of diabetic complications. Eleven of the 15 patients had retinopathy, six with proliferative changes; six had nephropathy; and six neuropathy. Most strikingly, four patients with anorexia nervosa developed acute painful polyneuropathy. In each case pain started when the eating disorder developed, almost coinciding with the peak of weight reduction. Remission of pain occurred as weight was regained. The symptoms were accompanied by abnormalities in peripheral nerve electrophysiology and autonomic nerve function, some improvements in which accompanied weight recovery. It is suggested that nutritional factors may contribute to the high rate of early onset diabetic complications, particularly neuropathy.     

Association of CNR1 and FAAH endocannabinoid gene polymorphisms with anorexia nervosa and bulimia nervosa: evidence for synergistic effects

Endocannabinoids modulate eating behavior; hence, endocannabinoid genes may contribute to the biological vulnerability to eating disorders. The rs1049353 (1359 G/A) single nucleotide polymorphism (SNP) of the gene coding the endocannabinoid CB1 receptor ( CNR1 ) and the rs324420 (cDNA 385C to A) SNP of the gene coding fatty acid amide hydrolase (FAAH), the major degrading enzyme of endocannabinoids, have been suggested to have functional effects on mature proteins. Therefore, we explored the possibility that those SNPs were associated to anorexia nervosa and/or bulimia nervosa. The distributions of the CNR1 1359 G/A SNP and of the FAAH cDNA 385C to A SNP were investigated in 134 patients with anorexia nervosa, 180 patients with bulimia nervosa and 148 normal weight healthy controls. Additive effects of the two SNPs in the genetic susceptibility to anorexia nervosa and bulimia nervosa were also tested. As compared to healthy controls, anorexic and bulimic patients showed significantly higher frequencies of the AG genotype and the A allele of the CNR1 1359 G/A SNP. Similarly, the AC genotype and the A allele of the FAAH cDNA 385C to A SNP were significantly more frequent in anorexic and bulimic individuals. A synergistic effect of the two SNPs was evident in anorexia nervosa but not in bulimia nervosa. Present findings show for the first time that the CNR1 1359 G/A SNP and the FAAH cDNA 385C to A SNP are significantly associated to anorexia nervosa and bulimia nervosa, and demonstrate a synergistic effect of the two SNPs in anorexia nervosa.     

Subjective Motives for Requesting In-Patient Treatment in Female with Anorexia Nervosa: A Qualitative Study

Background

Anorexia nervosa is a severe psychiatric disorder mainly affecting women. Its treatment is long and accepted with much difficulty, in particular in-patient treatment.

Aims

To describe the subjective motives of women with anorexia nervosa for requesting in-patient admission, from a qualitative analysis of application letters.

Methods

Participants were adult women (18 years and older) with anorexia nervosa who were admitted as in-patients in a referral hospital unit in France from January 2008 to December 2010. The application letters, prerequisites to admission, were studied by the interpretative phenomenological method of content analysis.

Results

63 letters have been analysed, allowing the identification of six themes related to requests for in-patient care: loss of control of behaviour, and of thoughts, mental exhaustion, isolation, inner struggle and fear of recovery.

Conclusions

Requests for in-patient admission were motivated by very personal, subjective experiences, unrelated to medical reasons for admission. These results may help improve pre-admission motivational work with individuals, by basing it on their subjective experience.     

Plasma adiponectin levels in women with anorexia nervosa.

Adiponectin is a plasma protein exclusively secreted by adipose tissue, which plays a role in modulating lipid and glucose metabolism. The plasma adiponectin concentration shows an inverse correlation with the body mass index in normal and obese individuals, but it has not been investigated in subjects with an extremely low body weight and undernutrition such as anorexia nervosa patients. We investigated plasma adiponectin levels in 21 females with anorexia nervosa. Nineteen healthy females served as the lean control group. The subjects with anorexia nervosa had a significantly lower weight and showed a tendency towards higher adiponectin levels than the control group. No correlation between adiponectin and BMI was found in patients with anorexia nervosa, while a linear negative correlation was seen in lean controls. The patient who showed the lowest adiponectin level reached a life-threatening state and required intravenous feeding in hospital. In association with improved nutrition and weight gain, the adiponectin level increased gradually until the body mass index was about 16 and then decreased subsequently as would be expected in lean normal subjects. These observations suggest that adipose tissue secretes less adiponectin and the adiponectin levels do not show an inverse correlation simply with body mass index in some subjects with severe undernutrition.     

Bulimia nervosa,binge eating,and psychogenic vomiting: a controlled treatment study and long term outcome.

An "epidemic" prevalence of binge eating and vomiting (bulimia nervosa) has been reported, and treatment has been claimed to be difficult. This paper describes a short term outpatient treatment programme of eclectic orientation capable of being conducted by non-specialist staff, under medical supervision, in local centres. The treatment programme was evaluated in a controlled trial and in long term follow up. In 30 women with severe bulimia the treatment programme significantly reduced their incidence of dietary manipulation without producing weight gain, weight disorder, or neurotic illness. After treatment all the women had fewer symptoms; 24 stopped binge eating and vomiting at the end of treatment, and a further four stopped shortly afterwards. During formal follow up 20 showed no dietary abuse and a further eight reduced their attacks to an average of three episodes a year: all judged treatment to be a success. Pretreatment indicators of poorer prognosis include alcohol abuse and a history of anorexia nervosa. Married patients experienced marital difficulties or illness in the spouse.     

Acute Gastric Dilatation in Anorexia Nervosa

Two patients with anorexia nervosa were treated on a general surgical unit for acute gastric dilatation. In both cases the dilatation rapidly followed an increase in the usual low dietary intake of the patients, and the ingestion of extra food may have initiated the acute episode. Conservative treatment with parenteral fluids, nasogastric intubation, and then a gradual return to a normal diet proved a satisfactory method of management. In one patient the anorexia itself was improved.