A genome-wide association study of anorexia nervosa.

Anorexia nervosa (AN) is a complex and heritable eating disorder characterized by dangerously low body weight. Neither candidate gene studies nor an initial genome-wide association study (GWAS) have yielded significant and replicated results. We performed a GWAS in 2907 cases with AN from 14 countries (15 sites) and 14 860 ancestrally matched controls as part of the Genetic Consortium for AN (GCAN) and the Wellcome Trust Case Control Consortium 3 (WTCCC3). Individual association analyses were conducted in each stratum and meta-analyzed across all 15 discovery data sets. Seventy-six (72 independent) single nucleotide polymorphisms were taken forward for in silico (two data sets) or de novo (13 data sets) replication genotyping in 2677 independent AN cases and 8629 European ancestry controls along with 458 AN cases and 421 controls from Japan. The final global meta-analysis across discovery and replication data sets comprised 5551 AN cases and 21 080 controls. AN subtype analyses (1606 AN restricting; 1445 AN binge-purge) were performed. No findings reached genome-wide significance. Two intronic variants were suggestively associated: rs9839776 (P=3.01 × 10(-7)) in SOX2OT and rs17030795 (P=5.84 × 10(-6)) in PPP3CA. Two additional signals were specific to Europeans: rs1523921 (P=5.76 × 10(-)(6)) between CUL3 and FAM124B and rs1886797 (P=8.05 × 10(-)(6)) near SPATA13. Comparing discovery with replication results, 76% of the effects were in the same direction, an observation highly unlikely to be due to chance (P=4 × 10(-6)), strongly suggesting that true findings exist but our sample, the largest yet reported, was underpowered for their detection. The accrual of large genotyped AN case-control samples should be an immediate priority for the field.

A change in objective sleep duration is associated with a change in the serum adiponectin level of women with overweight or obesity undergoing weight loss intervention.

BACKGROUND: Although the serum adiponectin level is inversely correlated to body mass index and closely associated with obesity and related diseases, neither the impact of weight loss on the adiponectin level nor other factors that might influence the adiponectin level during weight loss intervention are well documented. OBJECTIVE: The objective of the study is to assess the change in the serum adiponectin level during weight loss intervention and to determine if sleep parameters affect the serum adiponectin level. METHODS: Ninety women with overweight or obesity aged 25 to 65 years completed a 7-month cognitive behavioural therapy based weight loss intervention that included dieting, exercise and stress management. Serum adiponectin level, body fat percent, symptoms of depression and anxiety and objective sleep parameters, assessed by actigraphy, were measured at baseline and at the end of the intervention. RESULTS: The serum adiponectin level was significantly increased after the weight loss intervention (P < 0.001). In a multiple regression analysis, the change of the adiponectin level was positively associated with the magnitude of body fat loss (ß = -0.317, P < 0.001) and an increase of sleep minutes (ß = 0.210, P = 0.043). CONCLUSION: An increase in objective sleep duration was related to a significantly increased serum adiponectin level independently of the change of body fat during the weight loss intervention.

Changes in serum thyroid hormone, thyrotropin and thyroglobulin concentrations during thyroxine therapy in patients with solitary thyroid nodules.

We studied the effect of therapy with 0.1 mg/day T4 for 3 months on goiter size in 49 patients with solitary thyroid nodules. The nodule volume in 18 patients (responders) decreased by more than 50%. In this group the mean serum thyroglobulin (Tg) levels decreased significantly (from 425 to 61 micrograms/L; P less than 0.01). In the nonresponders the mean serum Tg levels did not change significantly (145 vs. 250 micrograms/L). The mean serum T4, free T4, free T3, and rT3 concentrations increased significantly in both groups during T4 therapy, serum T3 levels did not change, and serum TSH decreased. These findings demonstrate that serum Tg levels decrease when T4 therapy is effective. Thus, serum Tg measurements may prove a useful indicator of the efficacy of T4 therapy in patients with solitary nodules.

Insulin response to intravenous glucose in patients with anorexia nervosa showing low insulin response to oral glucose.

To investigate the cause of a low insulin secretory response to an oral glucose tolerance test (OGTT) in patients with anorexia nervosa (AN), we performed iv glucose tolerance tests (IVGTT) before and after treatment in 36 anorectic patients who showed low insulin secretion in response to the OGTT. These patients were subdivided into 3 groups by glucose tolerance curves during the OGTT: normal type, blood glucose level peaking 60 min or earlier after oral glucose ingestion; delayed type, blood glucose level peaking 90 min or later after oral glucose; and flat type, peak blood glucose level of 5.56 mmol/L or less after oral glucose. The results showed that the normal and flat type groups had normal glucose and insulin responses to iv glucose. In the delayed type group, in which the longest duration of AN before therapy was found, initial insulin secretion was decreased in response to both oral and iv glucose, indicating diminished pancreatic beta-cell function. After weight gain, this parameter improved significantly in both tests. The rate of glucose disappearance for the IVGTT was lower both before and after weight gain in this subgroup compared to that in normal controls, suggesting insulin resistance. In conclusion, the low insulin response to oral glucose seen in the flat type group may be due to the disturbance of gastrointestinal factors, such as motility. In contrast, the observations suggest that the delayed type group has beta-cell failure corrected by weight gain and has insulin resistance requiring a longer recovery time; these abnormalities are related to the duration of AN.

Development of spontaneous hypothyroidism in patients with Graves' disease treated with antithyroidal drugs: clinical, immunological, and histological findings in 26 patients.

Graves' disease may result eventually in hypothyroidism in approximately 5-20% of patients. In a few such patients hypothyroidism was associated with TSH-blocking antibodies, but whether the frequency of TSH-blocking antibodies in such patients is as high as it is (21%) in patients with primary myxedema is not known. This study was undertaken to determine the presence of various immunoglobulins [TSH binding inhibitor immunoglobulins, thyroid-stimulating antibodies (TSAb), and TSH-blocking antibodies] in 26 patients with Graves' disease who developed hypothyroidism from 0.5-10 yr or more after discontinuation of antithyroid drug therapy. Eight of the 26 patients (31%) had TSH-blocking antibodies, 16 (61%) had TSAb, and 14 (54%) had thyroid hormone binding inhibitor immunoglobulins. Thyroid needle biopsies were performed in 9 patients. Three of 5 patients who had subclinical hypothyroidism had chronic lymphocytic thyroiditis, and all had positive TSAb titers. Three patients had the fibrous variant of chronic lymphocytic thyroiditis; their TSAb values were 902%, 431%, and 1290%. One patient had follicular hyperplasia. We conclude that TSH-blocking antibodies may account for hypothyroidism in approximately one third of patients with Graves' disease who were previously treated with antithyroid drugs, and that autoimmune thyroiditis is comparable for the hypothyroidism in the remaining two thirds of Graves' disease patients.

HLA and thyrotoxic periodic paralysis in Japanese patients.

Periodic paralysis (PP) is a well recognized although rare and peculiar complication of thyrotoxicosis, especially among Chinese and Japanese patients. The susceptibility to autoimmune thyroid disease has recently been reported to be strongly linked to certain immunogenetic factors, and increased frequency of certain HLA antigens has been found in patients with Graves' disease. This study was, therefore, undertaken to determine HLA haplotypes in Japanese men with thyrotoxic periodic paralysis (TPP). HLA typing in 35 TPP patients and 263 normal men and women demonstrated highly significant increases (P less than 0.01) in HLA-A2, Cw3, and DRw8 in the TPP patients. In comparing TPP patients with thyrotoxic men who did not have PP, the frequency of DRw8 antigen was 2.5-fold greater in patients with PP than in those without it (62.8% vs. 28.6%). The data suggest that the HLA-DRw8 gene itself may play a significant role in the susceptibility to TPP among Japanese men.

Treatment of methimazole-induced agranulocytosis using recombinant human granulocyte colony-stimulating factor (rhG-CSF).

Agranulocytosis, although extremely infrequent, is a serious complication of antithyroidal drug therapy in patients with hyperthyroidism. Presently, there is no specific therapy for this life-threatening complication, and recovery time is highly variable. Recently, recombinant human granulocyte colony-stimulating factor (rhG-CSF) was reported to be effective in shortening the recovery time of the neutropenia in patients undergoing chemotherapy. The present study was undertaken to determine the efficacy of rhG-CSF administration in patients with methimazole-induced (MMI) agranulocytosis. Thirty-four patients (7 males and 27 females, ages 16-68 yr) with MMI agranulocytosis were divided into 3 groups: group A (n = 11) was treated with antibiotics only; group B (n = 11) received antibiotics and dexamethasone, 8 mg/day; and group C (n = 12) was treated with antibiotics and im injections of rhG-CSF, 75 micrograms/day. Patients in groups A and B were studied retrospectively. When rhG-CSF became available, patients in group C were studied prospectively. Bone marrow sternal punctures were performed in all group C patients who were then divided into 2 subgroups according to the granulocyte to erythrocyte count ratio (G:E). Group C1 (n = 6) had a G:E ratio of less than 0.5, and group C2 (n = 6) had a ratio of more than or equal to 0.5. Recovery time in all groups was defined as the number of days required for the peripheral granulocyte count to be greater than 1.0 x 10(9)/L. There was no significant difference in recovery time between groups A and B: 10.1 +/- 2.2 and 12.3 +/- 1.9 days (mean +/- SE), respectively. P was not significant; the administration of dexamethasone proved to be ineffective in shortening the time for recovery from peripheral granulocytes. On the other hand, recovery time was significantly shorter in group C (6.8 +/- 1.2 days mean +/- SE) compared with groups A and B (P < 0.05). Group C2 recovered in 2.2 +/- 0.6 days whereas group C1 took much longer, 9.8 +/- 1.3 days (P < 0.001). There was a direct correlation between the G:E ratio and the peripheral leucocyte count, r = 0.806, P < 0.01. Furthermore, rhG-CSF significantly shortened recovery time when the peripheral granulocyte count was greater than 0.1 x 10(9)/L (group C2) compared with patients whose counts were less than 0.1 x 10(9)/L (group C1), 2.2 +/- 0.4 vs. 8.6 +/- 1.3 days, respectively (P < 0.001). These data indicate that administration of steroids is ineffective in shortening the duration of recovery in patients with MMI agranulocytosis.(ABSTRACT TRUNCATED AT 400 WORDS)

Effect of cholinergic muscarinic receptor blockade on human growth hormone (GH)-releasing hormone-(1-44)-induced GH secretion in anorexia nervosa.

The presence of hypothalamic disturbances affecting GH secretion in anorexia nervosa has been suggested, although a normal GH response to GH-releasing hormone (GHRH) administration has been shown in these patients. The present study was performed to investigate the role of acetylcholine in regulating GH secretion by using pirenzepine, which selectively blocks muscarinic cholinergic receptors. Paired tests were performed in nine anorexia nervosa patients (age +/- SEM, 19.1 +/- 1.2 yr; percent ideal body weight, -32.7 +/- 2.2%) and in six normal controls (20.1 +/- 0.3 yr; -3.1 +/- 1.8%). GHRH-(1-44) (1 microgram/kg) was infused iv with and without pirenzepine pretreatment (0.6 mg/kg, iv). Basal levels of GH were not different in anorexia nervosa compared to normal controls, whereas, somatomedin-C levels were significantly lower in anorexia nervosa patients. However, after pirenzepine administration, the GHRH-stimulated GH responses were completely blocked in normal controls, but not in anorexia nervosa patients. These results suggested that altered muscarinic cholinergic mechanism are involved in the modulation of GH secretion in patients with anorexia nervosa.

Lack of effect of thyroxine administration on elevated thyroid stimulating hormone receptor antibody levels in treated Graves' disease patients.

Increased levels of antibodies to TSH receptors are thought to be a major cause of active Graves' disease or recurrence following therapy. It was recently reported that T4 administration during antithyroid drug treatment for Graves' disease resulted in a significant decrease of TSH receptor antibodies compared to drug therapy alone. It is known that these antibodies may remain elevated long after patients become euthyroid, so a large number of patients whose antibodies remained significantly elevated after 1 year of methimazole therapy were evaluated in the study. A total of 330 Graves' disease patients were treated with methimazole for 1 year. TSH receptor antibody titers remained persistently elevated in 195 patients. Thirty-five randomly selected patients were continued on maintenance doses of methimazole for a second year, and 160 patients were treated with a combination of methimazole and thyroxine for a second year. T4 doses needed ranged from 75-100 micrograms/day to maintain serum-free T4 and free T3 within the normal range. After 6 months of combined therapy, 35 patients were found to have suppressed serum TSH levels. The patients were divided after 18 months into three groups: A, B, and C. Group C, consisting of 35 randomly selected patients (8 males and 27 females) whose ages ranged from 12-62 years and who had been maintained on methimazole alone, served as controls. Group B, whose serum TSH levels were suppressed after 6 months of combined therapy, consisted of 9 males and 26 females whose ages were 15-66 years. Group A, 35 randomly selected patients with normal serum TSH levels after methimazole and thyroxine therapy for 6 months, consisted of 8 males and 27 females whose ages were 10-63 years. TSH receptor antibody titers gradually decreased in all three groups with drug therapy, and there was no significant difference in the titers at corresponding times, i.e. 0, 1.0, 1.5, and 2.0 years. After treatment for 2.0 years, all patients of the three groups were followed for a further 12 months. Rates of recurrence among the above three groups were not significantly different during the observation period. In the present study, T4 administration in combination with antithyroidal drugs had no effect on levels of antibodies to TSH receptors and no effect on rates of recurrence. The reason for the discrepant results in the present study from previous reports is not known.

Vascular responses to angiotensin II in anorexia nervosa.

Anorexia nervosa (AN) patients have a tendency to develop renin-angiotensin-aldosterone (RAA) abnormalities caused by abnormal behaviors expressed over long periods of time. Short-term dietary sodium intake is a known modulator of blood pressure response to infused angiotensin II (A II) in normal subjects. Therefore AN patients and normal gender-matched and age-matched controls were studied for vascular responses to exogenous A II. Untreated AN patients needed significantly greater quantities of exogenous A II to raise diastolic blood pressure (DBP) to over 20 mmHg for 30 min compared with controls (12.1 +/- 0.47 versus 7.6 +/- 0.69 ng/kg/min, p < 0.01). The amount of A II required to raise DBP to over 20 mmHg in AN patients in tests before and after completion of treatment (4.2 +/- 0.33 months later) was significantly different (12.1 +/- 0.47 versus 8.1 +/- 0.25 ng/kg/min, p < 0.01). There was no significant difference between AN patients following treatment and controls. Our results indicate that it requires long time before decreased A II responsiveness caused by chronic sodium depletion normalizes in AN patients.

Alterations in lymphocyte subsets and pituitary-adrenal gland-related hormones during fasting.

We investigated changes in the immunoendocrine system during fasting. Ten hospitalized patients aged 14-46 y with psychosomatic disorders fasted for 7 or 10 d. Blood samples were collected before and on days 3 and 7 of the 7-d fasts. When fasting continued to 10 d, an additional sample was taken on day 10. We measured blood cellularity (white blood cells and total lymphocytes), the total number and percentage of lymphocyte subsets (CD2, CD3, CD4, CD8, and CD19), natural killer (NK) cell activity, cytokines (interleukin 1 beta, interleukin 2, interleukin 6, granulocyte-macrophage colony stimulating factor, tumor necrosis factor alpha, and interferon gamma), and soluble interleukin 2 receptors. Corticotropin, cortisol, and dehydroepiandrosterone sulfate (DHEAS) concentrations were also determined. Although the total number of lymphocytes decreased during fasting, NK cell activity increased significantly. Plasma cortisol and DHEAS concentrations also increased significantly whereas changes in corticotropin concentrations were not significant. The total number and percentage of CD4 cells decreased significantly during fasting but no other lymphocyte subsets changed significantly. The percentage of CD4 cells was negatively correlated with cortisol concentrations during fasting. No detectable changes occurred in cytokines or soluble interleukin 2 receptors during the study. All measured immunoendocrine values that changed during fasting returned to prefasting values during the refeeding period. These findings indicate that fasting affects immune variables such as T cell subsets and NK cell activity at least in part through changes in adrenal gland-related hormones.

Orexin-A and leptin change inversely in fasting non-obese subjects.

OBJECTIVE: Leptin, neuropeptide-Y (NPY) and orexin are peptides regulating energy metabolism and appetite control. NPY and orexin are mainly found in the central nervous system and they have also recently been found in the peripheral nervous system. We investigated how fasting affects changes in circulating concentrations of these peptides and their association with nutritional and metabolic parameters in humans. DESIGN AND METHODS: Ten non-obese female patients with psychosomatic disorders fasted for 7 or 10 days. Blood samples were collected at 0800 h before fasting, on the 3rd and 7th days during the fast (with an additional sample taken on the 10th day when the fasting continued for 10 days) and on the 3rd and 7th days of refeeding. We measured blood concentrations of orexin-A, NPY, leptin, adrenocorticotropin, cortisol, insulin, C-peptide, glucose, and beta-hydroxybutyrate. RESULTS: Body mass index and plasma leptin concentrations concomitantly and significantly decreased during fasting, whereas serum orexin-A concentrations significantly increased and were negatively correlated with plasma leptin concentrations. Plasma NPY concentrations decreased slightly but were not significantly different from the prefasting values, and no significant relationship with leptin or orexin-A was found. Orexin-A and leptin concentrations showed a significant inverse correlation with serum glucose, insulin, C-peptide, and beta-hydroxybutyrate concentrations. Only changes in plasma leptin concentrations showed a significant negative correlation with serum cortisol concentrations. All the measured indices which changed during fasting returned to the prefasting concentrations by the 7th day of refeeding. CONCLUSION: Peripheral orexin-A and leptin concentrations inversely change during fasting, which is significantly correlated with energy metabolism in humans.

No leucine(7)-to-proline(7) polymorphism in the signal peptide of neuropeptide Y in Japanese population or Japanese with alcoholism.

We have screened 200 Japanese workers and 105 Japanese patients with alcoholism for the mutation in the signal peptide of pre-pro-neuropeptide Y resulting in a substitution of proline for leucine at position 7. This polymorphism was reported in the Finnish and Dutch populations recently. None of our subjects displayed the mutation at this site. Therefore, this allele does not play any role in the development of alcoholism in the Japanese population.

5-HT2A promoter polymorphism is not associated with anorexia nervosa in Japanese patients.

Genetic factors have been implicated in playing a significant role in susceptibility to anorexia nervosa (AN). Among many candidate genes for AN, an association with the A allele of the -1438G/A polymorphism in the promoter region of the 5-HT2A receptor has been reported. However, these findings are controversial and all patients studied to date have been Caucasian. This study was designed to determine whether this association is reproducible in Japanese subjects. This case-control study of a cohort of 75 female Japanese AN sufferers and 127 normal female control subjects revealed no significant association between the 5-HT2A promoter polymorphism and AN. Thus, at least for Japanese subjects, the A-allele of the -1438G/A polymorphism in the promoter region of the 5-HT2A receptor gene does not contribute to a predisposition to AN.

Analysis of tumor necrosis factor-alpha gene promoter polymorphisms in anorexia nervosa.

Elevated plasma tumor necrosis factor-alpha (TNFalpha) levels and enhanced spontaneous TNFalpha release from peripheral blood mononuclear cells in patients with anorexia nervosa (AN) have been reported. TNFalpha activates the hypothalamic-pituitary-adrenal axis and reduces food intake, which is characteristic of eating disorders. Recently, three novel polymorphisms in the 5'-flanking region of the TNFalpha gene were reported at positions -1031 (T --> C substitution), -863 (C --> A) and -857 (C --> T). Differences in these alleles are reportedly related to altered TNFalpha-transcriptional promoter activity. Therefore, we performed a case-control association analysis to determine whether any of those three polymorphisms in the TNFalpha promoter region were involved in a predisposition to AN. The results of our analysis of a cohort of 79 female Japanese AN sufferers and 127 normal female control subjects provide no support for the hypothesis that -1031T/C, -863 C/A and -857C/T polymorphisms in the TNFalpha gene promoter region influence the susceptibility to AN.

Responses of growth hormone and cortisol to intravenous glucose loading test in patients with anorexia nervosa.

The hypothalamic satiety and hunger centers appear to be affected by changes in circulating blood glucose concentrations. The response of the centers, in turn, is reflected by alterations in growth hormone (GH) and cortisol levels. There are no studies attempting to relate blood glucose and GH and cortisol changes in patients with anorexia nervosa (AN) during an intravenous glucose tolerance test (IVGTT). In the present inquiry, IVGTT (10 g) were performed on AN patients to characterize the satiety and hunger centers' responses to changes in glucose and insulin levels as reflected by GH and cortisol levels. Study participants were 15 female AN patients and eight healthy female volunteers. No significant differences in blood glucose levels were observed between the two groups. However, immunoreactive insulin (IRI) levels in AN patients were significantly lower than those in the control group. Although GH and cortisol concentrations were significantly suppressed after the infusion in the control group, the AN patients' GH levels paradoxically increased, and cortisol levels did not change. Moreover, a negative correlation was observed between delta GH and delta IRI in all individuals in this study (r = -.61, P less than .01). In conclusion, abnormal GH and cortisol responses to a 10-g IVGTT were found in patients with AN. delta GH levels correlated negatively with delta IRI levels. These data suggest that hypothalamic satiety and hunger centers in AN respond abnormally to change in blood glucose levels.

Effect of PACAP on gastric acid secretion in rats.

Pituitary adenylate cyclase-activating polypeptide (PACAP) is a new VIP-like brain-gut peptide. Its effects on the motility and secretory functions of the gastrointestinal system have been shown in previous studies. In this study we investigated the effect of intravenous PACAP on gastric acid secretion in conscious pylorus-ligated rats and in gastric fistula rats. PACAP showed significant inhibitory effects on pentagastrin- and histamine-stimulated gastric acid secretion, but no effect on basal or carbachol-stimulated secretion in pylorus-ligated rats. It did show dose-related inhibitory effects both on basal gastric acid secretion and on secretion stimulated by pentagastrin, histamine, or carbachol in gastric fistula rats. PACAP did not alter serum gastrin levels. Inhibition of prostaglandin synthesis with indomethacin and immunoneutralization of somatostatin with anti-somatostatin serum did not prevent the inhibitory effect of PACAP on gastric acid secretion in pylorus-ligated rats. We conclude that PACAP most likely has a direct effect on parietal cells and that this effect may be mediated, at least partially, by inhibition of the action of histamine on parietal cells.

Pituitary adenylate cyclase activating polypeptide (PACAP) can cross the vascular component of the blood-testis barrier in the mouse.

Pituitary adenylate cyclase activating polypeptide (PACAP), present in highest concentrations in the hypothalamus and testes, affects the release of LH, FSH, and prolactin, as well as Sertoli cell function. We examined the ability of the 38-amino acid form of PACAP labeled with 125I (I-P38) to cross the vascular component of the blood-testis barrier. The unidirectional influx constant (Ki) was 4.23 x 10(-3) ml/g-minute, which is about 5 times faster than the entry of LH and about 17 times faster than that of serum albumin. Entry occurred in part by a saturable transport system, with 20 micrograms/mouse of unlabeled P38 inhibiting transport by 40%. An analog of peptide T, which like PACAP is related to vasoactive intestinal polypeptide and has been found to have its own saturable transport system into the brain, did not alter the uptake of I-P38 by the testes. A dose of 10 micrograms/mouse, but not of 20 micrograms/mouse, was associated with a contraction of the vascular space of the testes. HPLC confirmed that a small but persistent percentage of the radioactivity recovered from the testes represented intact I-P38. These results suggest that circulating P38 may contribute to the testicular pool of PACAP, which may play an active role in the function of the testes.

Central and peripheral catecholamines regulate the exercise-induced elevation of plasma interleukin 6 in rats.

Several recent reports indicate that exercise elevates the plasma interleukin 6 levels; however, the precise regulation of such an elevation still remains to be clarified. In this study, in order to clarify the requirements of central and peripheral catecholaminergic system for this exercise-induced interleukin 6 elevation, rats were either intraperitoneally or intracerebroventricularly injected with 6-hydroxydopamine which depletes the catecholamine in the central or peripheral tissues. As a result, our exercise protocol elevated the plasma interleukin 6, ACTH, and corticosterone levels in response to exercise. All such exercise-induced increases in the interleukin 6, ACTH, and corticosterone levels were significantly inhibited by pretreatment with an intracerebroventricular injection of 6-hydroxydopamine. In the intraperitoneal 6-hydroxydopamine-treated animals, the exercise-induced interleukin 6 elevation was significantly suppressed compared with the vehicle-treated animals, although no significant difference was found in either the ACTH level or the corticosterone level between both groups of animals. These results thus suggest that central and peripheral catecholamines are involved in the regulation of the exercise-induced interleukin 6 elevation.

Differences between bulimia nervosa and binge-eating disorder in females with type 1 diabetes: the important role of insulin omission.

This study explored the differences between bulimia nervosa ("BN," n=22) and binge-eating disorder ("BED," n=11) in type 1 diabetic females and the factors most predictive of poor glycemic control in patients suffering from these disorders. These two groups and a control group without eating disorders (n=32) were compared across a number of demographic, psychological, and medical variables. BN manifested significantly more severe disturbances related to eating disorders, depression, anxiety, a higher rate of co-occurring mental disorders, and poorer psychosocial functioning compared with BED. BN also showed poorer glycemic control. Multivariate analysis indicated that higher serum glycosylated hemoglobin (HbA1c) levels were most associated with the presence of severe insulin omission in type 1 diabetic females with binge eating. Clinicians may be able to determine the psychological/medical severity of illness in these patients by identifying the presence of compensatory behaviors to prevent weight gain such as severe insulin omission, as described in the DSM-IV.