0021-972X/00/$03.00/0 The Journal of Clinical Endocrinology & Metabolism Copyright © 2000 by The Endocrine Society
Vol. 85, No. 7 Printed in U.S.A.
Plasma Leptin Response to Acute Fasting and Refeeding in Untreated Women with Bulimia Nervosa PALMIERO MONTELEONE, FRANCESCA BORTOLOTTI, MICHELE FABRAZZO, ANTONIETTA LA ROCCA, ANTONIO FUSCHINO, AND MARIO MAJ Institute of Psychiatry, University of Naples SUN, 80138 Naples, Italy ABSTRACT Leptin is known to regulate body weight, energy balance, and reproduction. Therefore, investigation of its physiology is of obvious interest in bulimia nervosa (BN), an eating disorder characterized by body weight-related psychopathology, acute changes in the energy balance, and reproductive alterations. To date, the few studies that have assessed leptin production in BN have had several limitations, including the measurement of blood leptin levels in treated patients and the lack of normal weight healthy controls, so that the information they provide is not conclusive. As the investigation of leptin dynamics is likely to be more informative, we decided to assess leptin response to acute fasting and refeeding in both untreated patients with BN and healthy controls. Twelve women meeting the diagnostic criteria for BN of the Diagnostic and Statistical Manual of Mental Disorders, and 10 healthy women of the same age range participated in a 3-day study. At 1800 h on day 1, they received a meal of 1088 Cal, with 53% carbohydrates, 17% protein, and 30% fat. Then, they fasted until 1800 h on day 2, when they received the same meal. On day 3, they received a standard hospital diet of 2600 Cal, divided into 3 meals, with the same percentages of nutrients as described above. Blood samples were col-
L
EPTIN, THE product of the ob gene, is a recently discovered protehormone that is thought to play a key role in the regulation of body weight (BW) and reproduction. Administration of exogenous leptin in mice not only causes weight loss, by suppressing food intake and increasing energy expenditure (1), but also reverses the fasting-induced inhibition of gonadotropin secretion (2) and the infertility of the ob/ob strain (3) as well as accelerates the onset of puberty in young female animals (4). Under conditions of unrestrained and timely food intake, leptin represents a static index of the available energy stored in the fat tissue. Under conditions of acute changes in caloric intake, leptin secretion varies independently of the available fat stores and serves as a sensor of energy balance (5). Moreover, circulating leptin does not merely reflect adiposity or energy balance, but is also affected by dietary macronutrient content (6). Given this background, investigation of leptin physiology is of obvious interest in patients with bulimia nervosa (BN). Indeed, BN is an eating disorder characterized by eating-related psychopathology resulting in episodes of massive food ingestion, prolonged periods of starvation, Received November 17, 1999. Revision received March 17, 2000. Accepted March 29, 2000. Address all correspondence and requests for reprints to: Palmiero Monteleone, M.D., Institute of Psychiatry, University of Naples SUN, Largo Madonna delle Grazie, 80138 Naples, Italy. E-mail:
[email protected].
lected at different time points for plasma leptin, glucose, and insulin measurements. In bulimic patients, plasma leptin values were significantly lower than in healthy women (P ⬍ 0.0001) and were positively related to body weight, expressed as body mass index (r ⫽ 0.86; P ⬍ 0.0001). The leptin response to the fasting/refeeding paradigm significantly differed between patients and controls (time ⫻ group interaction, P ⬍ 0.0001). In fact, in healthy subjects, acute fasting induced a 58% decline in the plasma leptin concentration, whereas such a decrease was only 7% in bulimic women (P ⬍ 0.001). After acute refeeding, plasma leptin increased in both groups, although in the patients it did not reach the absolute values observed in normal controls. No significant difference was observed between bulimics and controls in plasma insulin response to the fasting/refeeding paradigm, whereas an abnormal increase in blood glucose levels was observed in the patients after the first meal following acute fasting. We conclude that in untreated women with BN, leptin, despite its very low plasma values, still holds its function as a sensor of body weight changes, but loses its role of signaling acute changes in energy balance. (J Clin Endocrinol Metab 85: 2499 –2503, 2000)
and other compensatory behaviors aiming to reduce energy intake or increase its expenditure, generally with no pathological change in BW. Moreover, an imbalance of the micro/macronutrient composition of the diet as well as amenorrhea or oligomenorrhea frequently occur in women with this disorder. Therefore, leptin physiology could be deeply affected in BN, and, in turn, alterations of leptin production may be involved in the pathogenesis and/or the maintenance of the altered eating behavior of bulimic patients, because leptin functions as a satiety signal in the brain (7). At present, investigations of leptin secretion in BN are scanty. The only published studies explored morning plasma concentrations of leptin in bulimic women without comparison to normal controls (8 –10) or with the inclusion of patients at different stages of nutritional and behavioral treatment (11). Therefore, the information derived from these investigations cannot be conclusive. We thought that the assessment of leptin dynamics in drug-free bulimic patients might be more informative. Therefore, in the present study we investigated the leptin response to acute fasting and refeeding in both untreated women with BN and age-matched normal controls. Blood concentrations of glucose and insulin were also measured in this fasting and refeeding paradigm, because these substances have been claimed to modulate leptin production by adipose tissue (12–14).
2499
2500
JCE & M • 2000 Vol 85 • No 7
MONTELEONE ET AL. Subjects and Methods
collected in tubes with lithium heparin as anticoagulant. Plasma was separated by centrifugation and was stored at ⫺20 C.
Subjects Twenty-two women were recruited for the study. They were 12 out-patients enrolled among those consecutively attending the Eating Disorder Center of our institute and 10 healthy controls. According to the criteria of the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (15), patients fulfilled the diagnosis of BN. Diagnostic assessment was made by a trained interviewer using the Composite International Diagnostic Interview (16). Bulimic patients were all of the purging subtype, with binge episodes always followed by self-induced vomiting; none of them abused of laxatives or exercised excessively. The mean ⫾ sd daily binge frequency was 2.0 ⫾ 1.7 episodes/day. One patient had a past positive history of anorexia nervosa, one had a concomitant diagnosis of major depression, and one had an axis II diagnosis of borderline personality disorder. At the time of the study, one patient was amenorrheic, three were oligomenorrheic, and the remaining subjects had normal regular menses. All patients had been drug free for more than 6 weeks, and none of them had taken fluoxetine in the past. Control women were mentally healthy as assessed by the Composite International Diagnostic Interview and had no positive family history of mental disorders as assessed by the Family History Research Diagnostic Criteria (17). They were regularly menstruating and had normal diets. Both patients and healthy volunteers had normal physical examinations, normal values of routine blood and urine tests, and normal electrocardiogram. Female controls and menstruating patients were tested in the follicular phase of their menstrual cycles (days 5–10 from menses). None of the subjects was taking oral contraceptives or had a past history of alcohol or drug abuse.
Protocol Subjects gave written informed consent before study participation. Both patients and healthy subjects underwent a psychopathological assessment by the Eating Disorder Inventory (18) and the Bulimic Investigation Test Edinburgh (19) to evaluate the eating-related psychopathology; and the Hamilton Depression Rating Scale (20) to measure concomitant depressive symptoms. At 0800 h on day 1 after an overnight fast, each subject underwent a first blood sample collection for baseline plasma hormone screening, which included measurements of leptin, 17-estradiol, PRL, insulin, and cortisol. Then, subjects were free to go home and were asked to come back at 1600 h. Patients were carefully instructed to not binge or purge at home, and their eating behavior was checked by a relative. At 1800 h, subjects received a meal of 1088 Cal, with 53% carbohydrates, 17% protein, and 30% fat. Then they fasted up to 1800 h on day 2, when they received the same meal. On day 3, they received a standard hospital diet of 2600 Cal divided into three meals (0800 h, 307 Cal; 1300 h, 1207 Cal; 1800 h, 1088 Cal) with the same percentages of nutrients as described above. Blood samples were collected at 200 h on the first day, and at 0800, 1300, 1600, 1800, and 2000 h on both days 2 and 3. On the occasions of the meals, blood samples were drawn always before eating. Blood was
Biochemical analyses Plasma insulin, PRL, and 17-estradiol levels were determined by an immunoradiometric method, using commercial kits purchased from Biochem Immunosystem (Milan, Italy). The lower detection limits were 2.1 pmol/L for insulin, 0.4 g/L for PRL, and 18.3 pmol/L for 17estradiol. Intra- and interassay coefficients of variation were 2.6% and 4.7% for insulin, 5.4% and 8.6% for PRL, and 4.3% and 3.2% for 17estradiol, respectively. Plasma cortisol concentrations were determined by a double antibody RIA method, using commercial kits purchased from Biochem Immunosystem. The lower detection limit was 27 nmol/L. Intra- and interassay coefficients of variation were less than 5% and 8%, respectively. Plasma leptin values were determined by a sandwich enzyme-linked immunosorbent assay, using a commercial kit purchased from Alexis Biochemicals (Laufelfingen, Switzerland). The sensitivity of the method was 0.2 ng/mL; intra- and interassay coefficients of variation were 6.1% and 8.5%, respectively. Plasma glucose was determined by a commercial enzymatic UV method (Sigma, St. Louis, MO).
Statistics The BMDP statistical software package (21) was used for data analysis. Student’s t test for unpaired data was employed to analyze differences between patients and controls in both baseline hormone levels and clinical and demographic characteristics. Hormonal differences between the two groups in the fasting/refeeding paradigm were analyzed by two-way ANOVA with repeated measures, followed by the post-hoc Tukey’s test. The Pearson’s product-moment correlation test was employed to analyze possible correlations among the variables.
Results Clinical, demographic, and baseline hormonal data
No significant difference emerged between bulimic patients and control women in age, BW and body mass index (BMI; Table 1). Instead, as expected, bulimics scored significantly higher than healthy subjects on Eating Disorder Inventory, Bulimic Investigation Test Edinburgh, and Hamilton Depression Rating Scale (Table 1). In women with BN, morning plasma levels of leptin, 17estradiol, and PRL were significantly decreased, whereas plasma levels of cortisol, insulin, and glucose did not significantly differ from those in healthy controls (Table 1). A significant positive correlation was found between circulating leptin and BMI in both the patient group (r ⫽ 0.86;
TABLE 1. Clinical, demographic, and hormonal data of the study sample Bulimic women
Age (yr) BW (kg) Body mass index (kg/m2) Duration of the illness (yr) EDI total score BITE total score HDRS total score Plasma leptin (ng/mL) Plasma 17-estradiol (pmol/L) Plasma PRL (g/L) Plasma cortisol (nmol/L) Plasma insulin (pmol/L) Plasma glucose (mmol/L)
24.7 ⫾ 3.6 54.1 ⫾ 10.5 20.2 ⫾ 3.5 6.0 ⫾ 2.5 95.2 ⫾ 35.8 36.0 ⫾ 7.8 9.6 ⫾ 8.4 2.2 ⫾ 1.3 157.4 ⫾ 191.2 10.3 ⫾ 7.3 320 ⫾ 291 29.4 ⫾ 15.7 4.46 ⫾ 0.6
Healthy women
P
25.4 ⫾ 5.0 57.3 ⫾ 4.9 21.7 ⫾ 1.9
NS NS NS
21.5 ⫾ 14.6 4.1 ⫾ 2.9 1.4 ⫾ 2.7 15.0 ⫾ 10.0 307.9 ⫾ 207.7 19.5 ⫾ 5.4 330 ⫾ 141 48.8 ⫾ 25.1 4.92 ⫾ 0.4
⬍0.0001 ⬍0.0001 ⬍0.005 ⬍0.0001 ⬍0.05 ⬍0.002 NS NS NS
Values are the mean ⫾ SD. EDI, Eating Disorder Inventory; BITE, Bulimia Investigation Test Edinburgh; HDRS, Hamilton Depression Rating Scale.
LEPTIN DYNAMICS IN BULIMIA NERVOSA
P ⬍ 0.0001) and the healthy women (r ⫽ 0.67; P ⬍ 0.01). No other significant correlation emerged between leptin plasma levels and the plasma concentrations of remaining hormones or clinical and demographic variables in each group. Fasting and refeeding paradigm
Plasma leptin. Two-way ANOVA with repeated measures showed significant effects for group (F1,20 ⫽ 21.41; P ⬍ 0.0003) and time (F10,200 ⫽ 14.20; P ⬍ 0.0001) and a significant group ⫻ time interaction (F10,200 ⫽ 6.76; P ⬍ 0.0001), indicating that circulating leptin changed differently in the two groups throughout the fasting/refeeding paradigm. Indeed, compared with healthy women, plasma leptin concentrations were significantly lower in patients with BN at each time point of the study period (Fig. 1, top). Moreover, in control subjects, plasma leptin levels progressively dropped throughout the fasting period, reaching a nadir at 2000 h, then they progressively increased during normal refeeding. In patients with BN, plasma leptin concentrations decreased
2501
only slightly during acute fasting, whereas they increased after refeeding, even if they did not reach the values of normal controls. The 0800 h plasma leptin levels on days 2 and 3 were measured after the same period of fasting, and these values were highly interrelated (r ⫽ 0.96; P ⬍ 0.0001). We assumed them to be baseline values, and fasting- and refeedinginduced changes in circulating leptin were expressed as percentages of these values (Fig. 2). During the starvation period, two-way ANOVA with repeated measures showed significant effects for group (F1,20 ⫽ 11.33; P ⬍ 0.004) and time (F4,80 ⫽ 14.72; P ⬍ 0.0001) and a significant group ⫻ time interaction (F4,80 ⫽ 8.75; P ⬍ 0.0001); in the refeeding period, instead, no significant effect for group (F1,20 ⫽ 4.06; P ⫽ 0.06) and no significant group ⫻ time interaction (F4,80 ⫽ 0.86; P ⫽ 0.5) emerged. Indeed, in healthy women, a 58% drop in circulating leptin was observed at the end of the fasting period, whereas such a drop was only 7% in BN patients (Fig. 2, left). During refeeding, the percent increase in circulating leptin did not differ significantly in the two groups (Fig. 2, right). Plasma glucose. Two-way ANOVA with repeated measures showed no significant effect for group (F1,20 ⫽ 0.23; P ⫽ 0.6), but a significant effect for time (F10,200 ⫽ 12.74; P ⬍ 0.0001) and a significant group ⫻ time interaction (F10,200 ⫽ 6.10; P ⬍ 0.0001). During the fasting period, although a mild decrease in plasma glucose was observed in healthy women, glucose levels did not differ significantly between the two groups (Fig. 1, middle). After the first meal following the starvation period, blood levels of glucose promptly increased; this increase was significantly higher in bulimic patients than in healthy controls. During the refeeding day, no significant difference in plasma glucose was observed between the two groups. Plasma insulin. Two-way ANOVA with repeated measures showed no significant effects for group (F1,20 ⫽ 1.82; P ⫽ 0.1) and no significant group ⫻ time interaction (F10,200 ⫽ 0.85;
FIG. 1. Plasma leptin (top), glucose (middle), and insulin (bottom) responses to acute fasting and refeeding with normal diet in patients with BN and in healthy women. Data are expressed as the mean ⫾ SE. *, P ⬍ 0.0001; **, P ⬍ 0.005 (vs. correspondent time points in healthy controls, by Tukey’s test).
FIG. 2. Response of plasma leptin to acute fasting (left) and refeeding with normal diet (right) in bulimic patients and healthy women. Data represent the mean ⫾ SE) percent changes from 0800 h leptin concentrations. *, P ⬍ 0.02; **, P ⬍ 0.01; ***, P ⬍ 0.001 (vs. correspondent time points in bulimic patients, by Tukey’s test).
2502
MONTELEONE ET AL.
P ⫽ 0.5), but a significant effect for time (F10,200 ⫽ 71.00; P ⬍ 0.0001). In both subject groups, plasma insulin levels slightly dropped during the fasting period and promptly reincreased after the first meal, with no significant differences between bulimics and controls. On the refeeding day, plasma insulin levels did not differ significantly between the two groups (Fig. 1, bottom). Correlations
The areas under the curve from 0800 –2000 h were calculated for leptin, glucose, and insulin, separately for fasting and refeeding periods. Pearson’s correlation test did not show any significant correlation between the area under the curve of leptin and those of glucose or insulin in both fasting and refeeding conditions. Moreover, in both patients and healthy controls, no significant correlation emerged between 0800 h plasma leptin concentrations and its maximum percent decrease in response to acute fasting. Discussion
To the best of our knowledge, this is the first dynamic investigation of leptin physiology in patients with BN. The first major finding of this study is that in untreated bulimic women, plasma leptin levels are dramatically decreased compared to those in normal women matched for age and weight. Previous studies exploring leptin production in BN lacked healthy controls or used underweight healthy controls (8 –10) or included patients at different stages of nutritional and behavioral treatment (11); hence, their results are not directly comparable to ours. It has been claimed that circulating leptin reflects primarily BW changes and is directly related to the amount of fat stores, being high in obese individuals (22) and low in both underweight normal women (23) and underweight anorexic patients (8 –11, 24). Although in the present study we did not measure body fat, which may have differed between groups despite their similar BW and BMI, it is worth mentioning that within the same gender and age range it has been proved that BMI represents average body fat estimates (25). Therefore, although differences in body fat composition between groups cannot be completely excluded, on the basis of the present findings it seems possible to suggest that BW and fat stores are not the only determinant of leptin production, as our bulimic patients had very low plasma values of leptin despite no significant decrease in their BW and BMI. Hence, it can be argued that factors other than BW changes may be involved in the modulation of leptin production in BN. The second major finding of our study is that in bulimic patients, the secretion of leptin in response to acute fasting is almost completely absent, whereas its response to normal refeeding, although similar in percentage to that in normal controls, is not sufficient to restore normal blood levels of the hormone. To the contrary, the insulin response to this paradigm is completely preserved, whereas plasma glucose levels show a clear-cut significant increase after the first meal following acute starvation despite a normal blood peak of insulin. This abnormal increase in blood glucose could be ascribed to a reduced insulin sensitivity, which has been claimed in BN patients by some, although not all, researchers
JCE & M • 2000 Vol 85 • No 7
(26 –28). However, it is not clear why such an abnormal glucose increase occurs only after the first meal that follows the starvation period. We have no explanation for this phenomenon. This is a secondary finding of our study that does not seem to affect leptin results, as we did not observe any correlation between the response of leptin and those of insulin and glucose to the fasting/refeeding paradigm in either bulimic patients or healthy controls. The potential mechanisms underlying leptin alterations in BN are not known. One hypothesis is that the disordered eating, particularly the chronic fasting-gorging behavior of bulimic women, lowers leptin secretion and impairs its sensitivity to acute changes in energy intake. In this regard, it is worth mentioning that restrained eating has recently been shown to be associated with low leptin levels in young women (29). Restrained eating occurs in patients with BN, although it is not associated with decreased BW. Restrained eaters restrict their energy intake to avoid weight gain, whereas bulimics, despite the large caloric ingestion during the binge episodes, regurgitate a relatively large amount of calories by vomiting and engage in prolonged starvation to reduce their daily caloric ingestion. A caloric intake below the normal requirements of a given individual, even with no loss of fat mass, may have an impact on leptin synthesis. Furthermore, it has been recently shown that in normal individuals, experimental binge-eating dramatically flattens the leptin diurnal rhythm even when total calories are appropriate for weight (30). Therefore, one would suggest that the abnormal eating pattern of BN patients is an important modulator of leptin physiology. To this regard, it is worth mentioning that plasma levels of leptin have been shown to correlate significantly with the eating behavior score in a cohort of patients with anorexia nervosa or BN (31). Our results show an unambiguous positive correlation between circulating leptin and BMI in patients with BN, suggesting that, although at significantly lower levels, plasma leptin still functions as an endogenous signal of the BW differences. Therefore, it seems possible that in BN the role of leptin as a peripheral signal of the available energy stores is preserved, whereas its function as an index of acute changes in the energy balance is lost. This may have important pathogenic implications mainly for the binge-eating behavior of bulimic patients. In fact, as leptin behaves as a satiety factor (7), it is possible to speculate that its reduced levels and its failure to reach normal blood values after refeeding could contribute to the binge-eating behavior of bulimics. Indeed, human laboratory studies suggested that bulimic patients have diminished satiety responses to meals (32) that, on the basis of our results, could be mediated by an impaired production of leptin. Future studies assessing the relationship between plasma leptin and satiety or hunger sensations in bulimic patients may help to confirm or discard this hypothesis. We did not find any correlation between leptin plasma levels and psychopathological indexes of BN patients. Although the lack of a statistical correlation does not exclude a possible link between two variables, the present findings strongly support the idea that eating-related psychopathology and depressive symptoms are not connected to leptin changes in patients with BN. However, given the pulsatility
LEPTIN DYNAMICS IN BULIMIA NERVOSA
and the circadian variability of leptin secretion, future studies with larger patient samples and with many more measurements of leptin plasma levels are needed to specifically address this issue. According to previous findings by both other researchers and our group (33–36), we observed decreased blood levels of 17-estradiol and PRL, but normal plasma values of glucose, insulin, and cortisol in the BN group. Although a reciprocal influence between leptin and these substances has been claimed in humans (12–14), we could not detect any correlation between circulating leptin or leptin response to fasting/refeeding paradigm and baseline levels of plasma glucose, 17-estradiol, PRL, and cortisol in both patients and controls. As a consequence, it seems possible to exclude that these metabolic and endocrine abnormalities are involved in the reduced production of leptin in BN. In conclusion, we have demonstrated that in normal weight untreated bulimic women, plasma leptin levels are dramatically decreased and do not respond to short-term fasting, whereas they increase after acute normal refeeding, although not reaching normal concentrations. At this very low plasma values, leptin still holds its function as a sensor of BW changes, but it loses its role of signaling acute changes in the energy balance. The determinants of these changes and their roles in the genesis and/or maintenance of the altered eating behavior of BN patients remain to be determined.
12. 13.
14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.
References 1. Caro JF, Sinha MK, Kolaczynski JW, Zhang PL, Considine RV. 1996 Leptin: the tale of an obesity gene. Diabetes. 45:1455–1462. 2. Ahima RS, Prabakaran D, Mantzoros C, et al. 1996 Role of leptin in the neuroendocrine response to fasting. Nature. 382:250 –252. 3. Chehab FF, Lim ME, Lu R. 1996 Correction of the sterility defect in homozygous obese female mice by treatment with human recombinant leptin. Nat Gentile. 12:318 –2320. 4. Ahima R, Dushay J, Flier S, Prabakaran D, Flier J. 1997 Leptin accelerates the timing of puberty in normal female mice. J Clin Invest. 99:391–395. 5. Kolacynski JW, Considine RV, Ohannesian J, et al. 1996 Responses of leptin to short-term fasting and refeeding in humans. Diabetes. 45:1511–1515. 6. Havel PJ, Townsend R, Chaump L, Teff K. 1999 High-fat meals reduce 24-h circulating leptin concentrations in women. Diabetes. 48:334 –341. 7. Hauner H. 1996 Adipocytes as source and target of hormones: recent developments in adipose tissue physiology. Eat Pat Weight Control. 5:2–5. 8. Kopp W, Blum WF, von Prittwitz S, et al. 1997 Low leptin levels predict amenorrhea in underweight and eating disordered females. Mol Psychiatry. 2:335–340. 9. Kopp W, Blum WF, Ziegler A, et al. 1998 Serum leptin and body weight in females with anorexia and bulimia nervosa. Horm Metab Res. 30:272–275. 10. Mathiak K, Gowin W, Hebebrand J, et al. 1999 Serum leptin levels, body fat deposition, and weight in females with anorexia or bulimia nervosa. Horm Metab Res. 31:274 –277. 11. Ferron F, Considine RV, Peino R, Lado IG, Dieguez C, Casanueva FF. 1997 Serum leptin concentrations in patients with anorexia nervosa, bulimia ner-
27. 28. 29. 30. 31. 32. 33. 34. 35. 36.
2503
vosa and non-specific eating disorders correlate with the body mass index but are independent of the respective disease. Clin Endocrinol (Oxf). 46:289 –293. Boden G, Chen X, Mozzoli M, Ryan I. 1996 Effect of fasting on serum leptin in normal human subjects. J Clin Endocrinol Metab. 81:3419 –3423. Elbers JMH, Asscheman H, Seidell JC, Frolich M, Meinders AE, Gooren LJC. 1997 Reversal of the sex difference in serum leptin levels upon cross-sex hormone administration in transsexuals. J Clin Endocrinol Metab. 82:3267–3270. Newcomer JW, Selke G, Melson AK, Gross J, Vogler GP, Dagogo-Jack S. 1998 Dose-dependent cortisol-induced increases in plasma leptin concentration in healthy humans. Arch Gen Psychiatry. 55:995–1000. American Psychiatric Association. 1990 Diagnostic and statistical manual of mental disorders, 4th ed. Washington DC: American Psychiatric Press. World Health Organization. 1990 Composite International Interview (CIDI): a) CIDI interview version 1.0, b) CIDI-user manual, c) CIDI training manual, d) CIDI computer programs. Geneva: WHO. Andreasen NC, Endicott J, Spitzer RL, Winokur G. 1977 The family history method using diagnostic criteria: reliability and validity. Arch Gen Psychiatry. 34:1229 –1235. Garner DL, Olmstead MP, Polivy J. 1983 Development and validation of a multidimensional eating disorders inventory for anorexia nervosa and bulimia. Int J Eat Dis. 2:15–35. Henderson M, Freeman CPL. 1987 A self-rating scale for bulimia: the BITE. Br J Psychiatry. 150:18 –24. Hamilton M. 1960 A rating scale for depression. J Neurol Neurosurg Psychiatry. 23:56 – 62. Dixon J. 1985 BMDP Statistical Software. Berkeley: California University Press. Considine RV, Sinha MK, Helman ML, et al. 1996 Serum immunoreactiveleptin concentrations in normal-weight and obese humans. N Engl J Med. 334:292–295. Laughlin GA, Yen SSC. 1997 Ipoleptinemia in women athletes: absence of a diurnal rhythm with amenorrhea. J Clin Endocrinol Metab. 82:318 –321. Grinspoon S, Gulick T, Askari H, et al. 1996 Serum leptin in women with anorexia nervosa. J Clin Endocrinol Metab. 81:3861–3863. Heitmann BL. 1990 Evaluation of body fat estimated from body mass index, skinfolds and impedance. A comparative study. Eur J Clin Nutr. 44:831– 837. Kiriike N, Nishiwaki S, Nagata T, et al. 1990 Insulin sensitivity in patients with anorexia nervosa and bulimia. Acta Psychiatry Scand. 81:236 –239. Taniguchi A, Nakai Y, Fukushima M, et al. 1997 Intravenous glucose tolerance test-derived glucose effectiveness in bulimia nervosa. Metabolism. 46:484 – 486. Raphael FJ, Rodin DA, Peattie A, et al. 1995 Ovarian morphology and insulin sensitivity in women with bulimia nervosa. Clin Endocrinol (Oxf). 43:451– 455. von Prittwitz S, Blum WF, Ziegler A, Scharmann S, Remschmidt H, Hebebrand J. 1997 Restrained eating is associated with low leptin levels in underweight females. Mol Psychiatry. 2:420 – 422. Taylor AE, Hubbard J, Anderson EJ. 1999 Impact of binge eating on metabolic and leptin dynamics in normal young women. J Clin Endocrinol Metab. 84:428 – 434. Nakai Y, Hamagaki S, Kato S, Seino Y, Takagi R, Kurimoto F. 1999 Leptin in women with eating disorders. Metabolism. 48:217–220. Walsh TB, Kissileff HR, Cassidy SM, Dantzic S. 1989 Eating behavior of women with bulimia. Arch Gen Psychiatry. 46:54 –58. Schreiber W, Schweiger U, Werner D, et al. 1991 Circadian pattern of large neutral amino acids, glucose, insulin, and food intake in anorexia nervosa and bulimia nervosa. Metabolism. 40:503–507. Monteleone P, Brambilla F, Bortolotti F, Ferraro C, Maj M. 1998a Plasma prolactin response to d-fenfluramine is blunted in bulimic patients with frequent binge episodes. Psychol Med. 28:975–983. Monteleone P, Brambilla F, Bortolotti F, La Rocca A, Maj M. 1998b Prolactin response to d-fenfluramine is blunted in people with anorexia nervosa. Br J Psychiatry. 172:438 – 442. Stoving RK, Hangaard J, Hansen-Nord M, Hagen C. 1999 A review of endocrine changes in anorexia nervosa. J Psychiatry Res. 33:139 –152.
