- Email: [email protected]
Circulating leptin in patients with anorexia nervosa, bulimia nervosa or binge-eating disorder: relationship to body weight, eating patterns, psychopathology and endocrine changes
Psychiatry Research 94 Ž2000. 121]129
Circulating leptin in patients with anorexia nervosa, bulimia nervosa or binge-eating disorder: relationship to body weight, eating patterns, psychopathology and endocrine changes Palmiero MonteleoneU , Antonio Di Lieto, Alfonso Tortorella, Nicola Longobardi, Mario Maj Institute of Psychiatry, Uni¨ ersity of Naples SUN, Largo Madonna delle Grazie, 80138 Naples, Italy Received 27 September 1999; received in revised form 21 February 2000; accepted 28 February 2000
Abstract A decreased production of leptin has been reported in women with anorexia nervosa ŽAN. and has been attributed merely to the patients’ reduced body fat mass. The extent to which eating patterns, purging behaviors, psychopathology and endocrine changes may contribute to the genesis of leptin alterations has not been deeply investigated. Therefore, we measured plasma levels of leptin, glucose and other hormones in three groups of eating disorder patients with different body weight ŽBW., eating patterns and purging behaviors. Sixty-seven women, 21 with AN, 32 with bulimia nervosa ŽBN., 14 with binge-eating disorder ŽBED. and 25 healthy females volunteered for the study. We found that circulating leptin was significantly reduced in AN and BN patients, but significantly enhanced in women with BED. In anorexics, plasma glucose was decreased, whereas plasma cortisol was enhanced; blood concentrations of 17b-estradiol and prolactin ŽPRL. were reduced in both AN, BN and BED patients. In all subject groups, a strong positive correlation emerged between plasma levels of leptin and the subjects’ BW or body mass index, but not between leptin and psychopathological measures, plasma glucose, cortisol, PRL and 17b-estradiol. Since leptin was reduced in both underweight anorexics and normal weight bulimics, but increased in overweight BED women, who compulsively binge without engaging in compensatory behaviors, we suggest that factors other than BW may play a role in the determination of leptin changes in eating disorders. Q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Anorexia nervosa; Binge-eating disorder; Bulimia nervosa; Leptin
U
Corresponding author. Tel.: q39-081-566-6517; fax: q39-081-566-6523. E-mail address: [email protected] ŽP. Monteleone. 0165-1781r00r$ - see front matter Q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 5 - 1 7 8 1 Ž 0 0 . 0 0 1 4 4 - X
122
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129
1. Introduction
Leptin, the recently discovered adipocyte hormone, is thought to be an important component of the physiological system controlling energy homeostasis. When injected in the experimental animal, leptin behaves as a satiety factor decreasing food intake, increasing caloric expenditure and reducing body weight ŽBW. ŽMantzoros and Moschos, 1998.. Therefore, leptin contributes to BW regulation through modulating feeding behavior and energy expenditure. Moreover, leptin affects several neuroendocrine axes ŽAhima et al., 1996. and is currently thought to be an important factor triggering the onset of puberty and regulating the menstrual cyclicity in humans ŽApter, 1997.. Anorexia nervosa ŽAN. and bulimia nervosa ŽBN. are psychiatric disorders characterized by abnormal eating patterns and purging behaviors, aiming to reduce BW in AN and to prevent its increase in BN, because of the patients’ pathological fear of becoming fat. Malnutrition-induced metabolic and hormonal abnormalities occur in both syndromes and amenorrhea is a key diagnostic symptom of AN, while oligomenorrhea or amenorrhea may be present in BN. Given this background, the investigation of leptin production is of obvious interest in patients with eating disorders. It has been recently shown that, in underweight anorexic patients, leptin plasma levels are severely reduced and correlate with decreased body mass index ŽBMI. and body fat content ŽFerron et al., 1997; Hebebrand et al., 1997; Mantzoros et al., 1997; Kopp et al., 1998; Mathiak et al., 1999.. During the recovery of BW, leptin concentrations progressively re-increase to reach even higher than normal blood values ŽHebebrand et al., 1997.. Therefore, the reduction of body fat mass and BW has been maintained to be the major determinant of the lowered leptin secretion in AN. Some studies have assessed leptin production in BN ŽFerron et al., 1997; Kopp et al., 1997, 1998; Mathiak et al., 1999; Nakai et al., 1999.. However, most of them lack healthy controls
ŽKopp et al., 1997, 1998; Mathiak et al., 1999., one includes patients under different stages of treatment ŽFerron et al., 1997., and one, although including normal controls and untreated patients, groups subjects on the basis of their eating behavior score and not by diagnostic categories ŽNakai et al., 1999.. Hence, the results they provide cannot be conclusive. Besides AN and BN, DSM-IV includes in appendix B Žcriteria sets and axes provided for further studies. the category of binge-eating disorder ŽBED., which is characterized by bingeing episodes and eating-related psychopathology, as in BN, without compensatory behaviors. Therefore, this disorder offers the opportunity to study binge-related alterations without the confounding effects of compensatory behaviors. To the best of our knowledge, only one study has assessed leptin production in obese patients with BED ŽKarhunen et al., 1997.. Therefore, at present, investigations of leptin production in patients with eating disorders are warranted. In the present study, we measured plasma levels of leptin and other hormones in healthy women and in three groups of patients, consisting of women with AN, women with BN and women with BED, and assessed possible relationships between this hormone and demographic, clinical, psychopathological and biochemical variables of eating disorder patients.
2. Methods 2.1. Subjects A total of 92 women were recruited for the study. They were 67 outpatients attending the Eating Disorder Center of our Institute and 25 healthy controls. According to DSM-IV criteria, 21 patients fulfilled the diagnosis of AN, 32 the diagnosis of BN, and 14 the diagnosis of BED. Diagnostic assessment was made by a trained interviewer using the Composite International Diagnostic Interview ŽCIDI. ŽWHO, 1987.. Of the AN women, 14 were exclusively food restrictors
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129
and seven occasionally binged with a frequency no greater than twice a week. At the time of the study, two anorexics had a concomitant diagnosis of obsessive-compulsive disorder and two of major depression. Patients with BN were all of the purging subtype, with binge episodes always followed by self-induced vomiting; three of these patients also abused laxatives and two more exercised excessively. Three had a positive history of AN, two had a concomitant diagnosis of major depression, and one of borderline personality disorder. All of the AN women and nine bulimics were amenorrheic; six bulimics were oligomenorrheic; the remaining patients had normal regular menses. At the time of the study, all patients had been drug-free for more than 6 weeks. Control women were mentally healthy as assessed by the CIDI and had no positive family history of mental disorders as assessed by the Family History Research Diagnostic Criteria ŽAndreasen et al., 1977.. 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 a normal electrocardiogram. Female controls and patients who were normally menstruating were tested in the follicular phase of their menstrual cycle Ždays 5]10 from menses.. None of the subjects were taking oral contraceptives or had a past history of alcohol or drug abuse.
123
after an overnight fast. Blood was collected in tubes with lithium heparin as anticoagulant. Plasma was separated by centrifugation and stored at y208C. 2.3. Biochemical analyses Plasma prolactin ŽPRL. and 17b-estradiol levels were determined by an immunometric method ŽMAIA clone., using commercial kits purchased from Biochem Immuno System ŽMilan, Italy.. The lower detection limits were 0.4 ngrml for PRL and 5 pgrml for 17b-estradiol. Intra- and interassay coefficients of variation were 5.4% and 8.6% for PRL, and 4.3% and 3.2% for 17b-estradiol. Plasma cortisol concentrations were determined by a double-antibody RIA method, using commercial kits purchased from Biochem Immuno System ŽMilan, Italy.. The lower detection limit was 27 nmolrl. Intra- and inter-assay coefficients of variation were less than 5% and 8%. Plasma leptin values were determined by a sandwich enzyme-linked immunosorbant assay, using a commercial kit purchased from Alexis Biochemicals ŽLaufelfingen, Switzerland.. The sensitivity of the method was 0.2 ngrml; intraand inter-assay coefficients of variation were 6.1% and 8.5%, respectively. Plasma glucose was determined by a commercial enzymatic UV method ŽSigma Diagnostics, St. Louis, MO, USA.. 2.4. Data analysis
2.2. Procedure Subjects gave written informed consent prior to study participation. Both patients and healthy subjects underwent a psychopathological assessment with the Eating Disorder Inventory ŽEDI. ŽGarner et al., 1983. and the Bulimic Investigation Test Edinburgh ŽBITE. ŽHenderson and Freeman, 1987. to evaluate eating-related psychopathology; and the Hamilton Depression Rating Scale ŽHDRS. ŽHamilton, 1960. to measure concomitant depressive symptoms. Each subject underwent a blood sample collection in the morning, between 8.00 and 9.00 h,
As there were significant deviations from normality in the data, non-parametric statistical analyses were used. Where the Kruskal]Wallis analysis of variance ŽANOVA. showed significant differences among the groups, the Mann]Whitney U-test was employed to assess differences between patients and controls. The Spearman rank order correlation test was used to examine the relation between leptin values, on the one hand, and demographic, clinical or hormonal data, on the other. In addition, a multiple regression analysis with a backward stepwise procedure was used to define the biochemical variables most predictive of leptin concentrations.
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129
124
3. Results
lated psychopathology and depression ŽTable 1.. No significant differences were observed in any psychometric measure between anorexics of the restricting vs. the binge-purging type.
3.1. Demographic and clinical data
Kruskal]Wallis ANOVA showed significant inter-group differences in BW Ž H s 55.457, d.f.s 3, P- 0.0001. and BMI Ž H s 60.286, d.f.s 3, P0.0001. and percent of ideal BW Ž H s 60.035, d.f.s 3, P- 0.0001., but not in age, height or duration of the illness ŽTable 1.. With respect to healthy women, AN patients had significantly lower BW Ž z s 5.424, P- 0.0001, Mann]Whitney test., BMI Ž z s 5.711, P- 0.0001. and percent of ideal BW Ž z s 5.777, P - 0.0001., whereas patients with BED had significantly higher BW Ž z s 4.186, P - 0.0001., BMI Ž z s 4.098, P 0.0001. and percent of ideal BW Ž z s 4.274, P0.0001.. No significant difference in these variables was observed between bulimics and healthy women and between anorexics of the restricted subtype and those of the binge-purging subtype. For the bulimic group, the mean bingervomiting frequency was 1.9" 1.3 episodesrday; for the BED group, the binge frequency was 2.7" 0.8 episodesrday. As compared with healthy women, anorexic, bulimic and BED patients had significantly higher scores on standardized assessments of eating-re-
3.2. Biochemical data
Kruskal]Wallis ANOVA showed significant inter-group differences in plasma levels of leptin Ž H s 60.848, d.f.s 3, P- 0.0001., 17b-estradiol Ž H s 22.868, d.f.s 3, P- 0.0001. and PRL Ž H s 35.814, d.f.s 3, P- 0.002., but not in plasma glucose Ž H s 5.596, d.f.s 3, P s 0.1. or cortisol concentrations Ž H s 6.520, d.f.s 3, Ps 0.08.. As compared with healthy women, AN and BN patients exhibited significantly decreased plasma levels of leptin, whereas women with BED had significantly enhanced blood concentration of the hormone ŽFig. 1.. No significant difference was observed in leptin plasma levels between anorexics of the restricting vs. the binge-purging type. Bulimics with amenorrhea or oligomenorrhea had leptin plasma values that were lower than values in the normally menstruating ones Ž3.16" 1.6 ngrml vs. 6.1" 4.8 ngrml., although the difference did not reach statistical significance. Plasma levels of 17b-estradiol and PRL were significantly reduced in all the patient groups
Table 1 Demographic and clinical characteristics of study sample a
Number of subjects Age Žyears. Body weight Žkg. Height Žcm. Body mass index Žkgrm2 . Percent of ideal body weight EDI total score BITE total score HDRS total score Duration of the illness Žyears.
Healthy women
Women with anorexia nervosa
Women with bulimia nervosa
Women with binge-eating disorder
25 25.4" 4.2 56.9" 6.1 161.7" 4.7 21.7" 2.2 101.2" 10.2 24.4" 15.8 5.7" 3.3 1.3" 2.8 Not applicable
21 25.0" 6.1 41.9" 5.6U 162.0" 5.0 16.0" 1.6U 74.4" 7.2U 72.1" 20.3U 19.9" 9.0U 16.9" 6.7U 5.3" 5.1
32 23.2" 3.3 53.6" 8.8 161.8" 7.7 20.5" 2.8 95.7" 21.4 96.1" 26.8U 36.6" 7.6U 12.6" 8.3U 4.9" 2.8
14 26.0" 6.9 82.2" 20.2U 163.5" 5.6 30.9" 7.9U 144.9" 35.8U 95.4" 22.9U 34.0 " 8.5U 11.9" 7.0U 4.1" 3.5
All data are presented as mean " S.D. U P- 0.0001 vs. healthy women ŽMann]Whitney test..
a
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129
125
Fig. 1. Plasma leptin concentrations in healthy women and in women with anorexia nervosa, bulimia nervosa and binge-eating disorder. Data are expressed as mean " S.E.M. U P- 0.005 and UU P- 0.0001 vs. healthy women ŽMann]Whitney test..
ŽTable 2.. No significant difference in these parameters was observed between anorexics of the restricting vs. the binge-purging type. Eumenorrheic bulimics had plasma levels of 17bestradiol which were higher than those of bulimics with menstrual irregularities Ž zs 1.71, P- 0.04. and not significantly different from those of healthy women. PRL plasma levels did not significantly differ between the two bulimic subgroups. Since in the AN group mean baseline plasma cortisol concentrations were higher and plasma glucose was lower than in healthy subjects, a further ANOVA was run to test for possible
differences between AN patients and control women that could have been missed in a fourgroup comparison. This analysis showed a significant difference between anorexic patients and healthy subjects for both cortisol and glucose values Ž H s 6.859, d.f.s 1, P- 0.01 and H s 5.260, d.f.s 1, P - 0.025, respectively.. The Mann]Whitney test confirmed that anorexics had higher plasma cortisol Ž z s 2.602, P- 0.005. and lower plasma glucose Ž z s 2.304, P- 0.02. concentrations than healthy women with no significant differences between the two AN subgroups. In order to evaluate whether differences in estradiol and cortisol levels between groups per-
Table 2 Comparison between healthy women and women with eating disorders for plasma hormone and glucose concentrations a
Plasma prolactin Žngrml. Plasma cortisol Žnmolrl. Plasma 17b-estradiol Žpgrml. Plasma glucose Žmgrdl.
Healthy women Ž n s 15.
Women with anorexia nervosa Ž n s 21.
Women with bulimia nervosa Ž n s 32.
Women with binge-eating disorder Ž n s 14.
19.0" 7.6 304 " 107 72.4" 59.2 93.6" 20.4
7.2" 2.7U 445 " 171UU 18.7" 16.1U 82.7" 11.5UUU
9.7" 6.5U 370 " 238 42.0" 40.5UU 84.2" 18.4
9.0" 3.9U 350 " 120 34.0" 8.5UUU 90.0" 15.3
All data are presented as mean " S.D. U P- 0.0001; UU P- 0.005; UUU P- 0.02 vs. healthy women ŽMann]Whitney test..
a
126
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129
sisted after adjustment for leptin plasma levels, an analysis of covariance was performed with leptin as the dependent variable. The results of this analysis showed that neither cortisol Ž F s 0.01, d.f.s 1, 43, Ps NS. nor 17b-estradiol Ž F s 0.44, d.f.s 1, 87, Ps NS. was significantly associated with leptin. 3.3. Correlations Leptin plasma levels were positively correlated to both BW and BMI in AN Žr s 0.44, P- 0.02; r s 0.54, P- 0.005, respectively., BN Žr s 0.31, P- 0.04; r s 0.36, P- 0.02. and BED patients Žr s 0.78, P- 0.0003; r s 0.81, P- 0.0001. as well as in healthy women Žr s 0.59, P- 0.0008; r s 0.62, P- 0.0004. ŽFig. 2.. The multiple regression analysis showed that glucose, 17b-estradiol, PRL and cortisol were not
significant determinants of leptin concentrations in our sample.
4. Discussion This is the first study to explore leptin production across the eating-disorder spectrum. We found that circulating morning leptin was significantly reduced in untreated patients with AN and in those with BN, whereas it was significantly increased in women with BED, compared with age-matched healthy females. Consistent with our findings, decreased blood levels of leptin in AN patients have been reported by other authors ŽFerron et al., 1997; Hebebrand et al., 1997; Mantzoros et al., 1997; Kopp et al., 1998; Mathiak et al., 1999., who suggested that the lowered production of the adipocyte hormone
Fig. 2. Relationships between plasma leptin concentrations and body mass index ŽBMI. in healthy women and in patients with anorexia nervosa, bulimia nervosa or binge-eating disorder ŽBED..
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129
merely reflects the reduced amount of body fat mass, which typically occurs in AN. Previous studies exploring leptin production in BN are not directly comparable with our study because of the lack of normal controls ŽKopp et al., 1997, 1998; Mathiak et al., 1999., the inclusion of patients at different stages of nutritional and behavioral treatment ŽFerron et al., 1997. or the grouping of patients on the basis of their eating behavior score and not by diagnostic categories ŽNakai et al., 1999.. In agreement with previous reports ŽFerron et al., 1997; Hebebrand et al., 1997; Mantzoros et al., 1997; Kopp et al., 1998; Mathiak et al., 1999., our results show an unambiguous positive correlation between circulating leptin and BW or BMI in all the subject groups. However, our results also suggest that the reduction in BW is not the only determinant of the lowered secretion of leptin in eating-disorder patients. Indeed, bulimic women, who have abnormal eating behavior, like the anorexic ones, did show a clear-cut decrease in circulating leptin in spite of the maintenance of a normal BW. This observation strongly indicates that, in addition to BW, other factors play a role in regulating leptin production in eating disorders. Restrained eating has recently been shown to be associated with low leptin levels in young women Žvon Prittwitz et al., 1997.. Restrained eating occurs in both patients with AN and those with BN, although in the latter it is not associated with decreased BW. Restrained eaters restrict their energy intake to avoid weight gain and bulimics, despite the large caloric ingestion during binge episodes, regurgitate a relatively large amount of calories by vomiting and engage in prolonged starvation in order to reduce their daily caloric ingestion. A caloric intake below the normal requirements of the respective individual, even with no loss of fat mass, may have an impact on leptin synthesis. Therefore, one would suggest that, besides BW, the eating pattern is important in the modulation of leptin secretion in humans. Although in our study we did not measure the eating behavior of our patients by ad hoc psychometric scales, it is worth mentioning that plasma levels of leptin have been shown to significantly
127
correlate with the eating behavior score in a cohort of patients with AN or BN ŽNakai et al., 1999.. Furthermore, in the present study, patients with BED, who compulsively binge, but do not fast or engage in compensatory behaviors, exhibited an increased concentration of plasma leptin as one would normally expect according to their enhanced BW. Therefore, it seems possible to conclude that low plasma leptin may be a sensitive indicator not only of the overall nutritional status, but also of metabolic changes occurring with restrained eating and purging behaviors. Since leptin behaves as a satiety factor, it is possible to speculate that its reduced levels in both AN and BN could contribute to the increased hunger sensation of the anorexic women ŽOwen et al., 1985. and to the binge-eating behavior of bulimics. Indeed, human laboratory studies suggested that bulimic patients have diminished satiety responses ŽWalsh et al., 1989. that, on the basis of our results, could be mediated by a decreased production of leptin. This may lead to binge eating. In women with BED, who exhibit increased levels of leptin, an impaired sensitivity to the enhanced circulating leptin may be involved in the pathogenesis andror maintenance of the bingeing behavior. In most cases, BED patients are truly obese, and a reduced sensitivity to the abnormally high circulating leptin levels has been hypothesized in human obesity ŽMantzoros and Moschos, 1998.. Future studies assessing the relationship between plasma leptin and satiety or hunger sensations may help to confirm or disconfirm this hypothesis. A further observation of our study is that, in the BN group, patients with amenorrhea or oligomenorrhea exhibited lower circulating leptin levels than those with regular menses, although this difference did not reach statistical significance. This finding confirms previous suggestions that reduced leptin may be an important determinant of the menstrual abnormalities occurring in patients with eating disorders ŽKopp et al., 1997., even if, due to the relatively small number of subjects in our subgroups of bulimic women, a type-II statistical error cannot be ruled out. According to previous findings by both other authors and our group ŽSchreiber et al., 1991;
128
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129
Monteleone et al., 1998a,b, 1999; Stoving et al., 1999., we observed decreased blood levels of glucose and increased concentrations of cortisol in AN women, and reduced plasma values of 17bestradiol and PRL in all the patient groups. It has been suggested that blood glucose may influence leptin secretion likely by inducing insulin production ŽBoden et al., 1996.. Insulin, in fact, is believed to modulate leptin synthesis by both enhancing leptin mRNA expression and exerting a trophic effect on the adipocytes ŽMantzoros and Moschos, 1998.. Similarly, estrogens and glucocorticoids have been reported to increase leptin production in humans ŽElbers et al., 1997; Newcomer et al., 1998., and leptin, in turn, has been shown to regulate the circadian oscillations of both luteinizing hormone and estradiol in normal women ŽLicinio et al., 1998. and to inhibit the production of cortisol ŽBornstein et al., 1997., which suggests the existence of feedback loops between leptin and both ovarian and adrenal steroids. Therefore, one could speculate that, in our eating-disorder patients, the changes in plasma levels of leptin, glucose, 17b-estradiol and cortisol may be mutually linked. However, on the one hand, we found that leptin did not affect the differences in 17b-estradiol and cortisol levels among the groups, and, on the other hand, neither the hormonal parameters nor glucose plasma levels proved to be a significant determinant of leptin plasma concentrations. Therefore, it seems possible to exclude that glucose and endocrine abnormalities are linked to the altered leptin secretion in eating disorder patients. Some limitations of the present study need to be discussed. First of all, it must be noted that body fat correlates with leptin levels to a higher degree than does BW, mainly in patients with eating disorders ŽMathiak et al., 1999.. Hence, a shortcoming of the present study is that we did not measure body fat, which may have differed between groups independently of BMI. However, within the same gender and age range, it has been shown that BMI may represent average body fat estimates ŽHeitmann, 1990.. A further limitation of our study is that our findings refer to a single morning measurement of blood hormones in spite of the well-recognized
circadian rhythmicity in the secretion of these substances. However, although we cannot exclude that phase shifts in the circadian rhythms may have affected our results, it is of note that the present findings are in line with previously reported values of morning plasma leptin, cortisol and PRL in eating-disorder patients. Finally, the measurement of blood levels of insulin, insulinlike growth factor I and triiodothyronine would have been of interest, since these hormones have been linked to leptin production in eating-disorder patients ŽGrinspoon et al., 1996; Nakai et al., 1999.. In conclusion, circulating leptin is reduced in untreated underweight anorexic patients and normal weight bulimics, whereas it is increased in overweight BED patients. BW, menstrual status, eating patterns and purging behaviors, but not psychopathology or other metabolic and endocrine abnormalities, seem to be the major determinants of leptin changes in eating disorders. References Ahima, R.S., Prabakaran, D., Mantzoros, C., Qu, D.Q., Lowell, B., Maratosflie, E., Flier, J.S., 1996. Role of leptin in the neuroendocrine response to fasting. Nature 382, 250]252. Andreasen, N.C., Endicott, J., Spitzer, R.L., Winokur, G., 1977. The family history method using diagnostic criteria: reliability and validity. Archives of General Psychiatry 34, 1229]1235. Apter, D., 1997. Leptin in puberty. Clinical Endocrinology 47, 175]176. Boden, G., Chen, X., Mozzoli, M., Ryan, I., 1996. Effect of fasting on serum leptin in normal human subjects. Journal of Clinical Endocrinology and Metabolism 81, 3419]3423. Bornstein, S.R., Uhlmann, K., Haidan, A., Ehrhart-Bornstein, M., Scherbaum, W.A., 1997. Evidence for a novel peripheral action of leptin as a metabolic signal to the adrenal gland: leptin inhibits cortisol release directly. Diabetes 46, 1235]1238. Elbers, J.M.H., Asscheman, H., Seidell, J.C., Frolich, M., Meinders, A.E., Gooren, L.J.C., 1997. Reversal of the sex difference in serum leptin levels upon cross-sex hormone administration in transsexuals. Journal of Clinical Endocrinology and Metabolism 82, 3267]3270. Ferron, F., Considine, R.V., Peino, R., Lado, I.G., Dieguez, C., Casanueva, F.F., 1997. Serum leptin concentrations in patients with anorexia nervosa, bulimia nervosa and nonspecific eating disorders correlate with the body mass index but are independent of the respective disease. Clinical Endocrinology 46, 289]293.
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129 Garner, D.L., Olmstead, M.P., Polivy, J., 1983. Development and validation of a multidimensional eating disorders inventory for anorexia nervosa and bulimia. International Journal of Eating Disorders 2, 15]35. Grinspoon, S., Gulick, T., Askari, H., Landt, M., Lee, K., Anderson, E., Ma, Z.M., Vignati, L., Bowsher, R., Herzog, D., Klibanski, A., 1996. Serum leptin levels in women with anorexia nervosa. Journal of Clinical Endocrinology and Metabolism 81, 3861]3863. Hamilton, M., 1960. A rating scale for depression. Journal of Neurology, Neurosurgery and Psychiatry 23, 56]62. Hebebrand, J., Blum, W.F., Barth, N., Coners, H., Englaro, P., Juul, A., Ziegler, A., Warnke, A., Rascher, W., Remschmidt, H., 1997. Leptin levels in patients with anorexia nervosa are reduced in the acute stage and elevated upon short-term weight restoration. Molecular Psychiatry 2, 330]334. Heitmann, B.L., 1990. Evaluation of body fat estimated from body mass index, skinfolds and impedance. A comparative study. European Journal of Clinical Nutrition 44, 831]837. Henderson, M., Freeman, C.P.L., 1987. A self-rating scale for bulimia: the BITE. British Journal of Psychiatry 150, 18]24. Karhunen, L., Haffner, S., Lappalainen, R., Turpeinen, A., Miettinen, H., Uusitupa, M., 1997. Serum leptin and shortterm regulation of eating in obese women. Clinical Sciences 92, 573]578. Kopp, W., Blum, W.F., von Prittwitz, S., Ziegler, A., Lubbert, H., Emons, G., Herzog, W., Herpertz, S., Deter, H-C., Remschmidt, H., Hebebrand, J., 1997. Low leptin levels predict amenorrhea in underweight and eating disordered females. Molecular Psychiatry 2, 335]340. Kopp, W., Blum, W.F., Ziegler, A., Mathiak, K., Lubbert, H., Herpertz, S., Deter, H.C., Hebebrand, J., 1998. Serum leptin and body weight in females with anorexia and bulimia nervosa. Hormone and Metabolic Research 30, 272]275. Licinio, J., Negrao, A.B., Mantzoros, C., Kaklamani, V., Wong, M.-L., Bongiorno, P.B., Mulla, A., Cearnal, L., Veldhuis, J.D., Flier, J.S., McCann, S.M., Gold, P.W., 1998. Synchronicity of frequently sampled, 24-h concentrations of circulating leptin, luteinizing hormone, and estradiol in healthy women. Proceedings of the National Academy of Sciences USA 95, 2541]2546. Mantzoros, C.S., Moschos, S.J., 1998. Leptin: in search of roles in human physiology and pathophysiology. Clinical Endocrinology 49, 551]567. Mantzoros, C., Flier, J.S., Lesem, M.D., Brewerton, T.D., Jimerson, D.C., 1997. Cerebrospinal fluid leptin in anorexia
129
nervosa: correlation with nutritional status and potential role in resistance to weight gain. Journal of Clinical Endocrinology and Metabolism 82, 1845]1851. Mathiak, K., Gowin, W., Hebebrand, J., Ziegler, A., Blum, W.F., Felsenberg, D., Lubbert, H., Kopp, W., 1999. Serum leptin levels, body fat deposition, and weight in females with anorexia or bulimia nervosa. Hormone and Metabolic Research 31, 274]277. 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. Psychological Medicine 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. British Journal of Psychiatry 172, 438]442. Monteleone, P., Brambilla, F., Bortolotti, F., Maj, M., 1999. Serotonergic dysfunction across the eating disorders: relationship to eating behaviour, purging behaviour, nutritional status and general psychopathology. Submitted. Nakai, Y., Hamagaki, S., Kato, S., Seino, Y., Takagi, R., Kurimoto, F., 1999. Leptin in women with eating disorders. Metabolism 48, 217]220. Newcomer, J.W., Selke, G., Melson, A.K., Gross, J., Vogler, G.P., Dagogo-Jack, S., 1998. Dose-dependent cortisol-induced increases in plasma leptin concentration in healthy humans. Archives of General Psychiatry 55, 995]1000. Owen, W.P., Kalmi, K.A., Gibbs, J., Smith, G.P., 1985. Satiety responses in eating disorders. Journal of Psychiatric Research 19, 279]284. Schreiber, W., Schweiger, U., Werner, D., Brunner, G., Tuschl, R.J., Laessle, R.G., Krieg, J.C., Fichter, M.M., Pirke, K.M., 1991. Circadian pattern of large neutral amino acids, glucose, insulin, and food intake in anorexia nervosa and bulimia nervosa. Metabolism 40, 503]507. Stoving, R.K., Hangaard, J., Hansen-Nord, M., Hagen, C., 1999. A review of endocrine changes in anorexia nervosa. Journal of Psychiatric Research 33, 139]152. von Prittwitz, S., Blum, W.F., Ziegler, A., Scharmann, S., Remschmidt, H., Hebebrand, J., 1997. Restrained eating is associated with low leptin levels in underweight females. Molecular Psychiatry 2, 420]422. Walsh, T.B., Kissileff, H.R., Cassidy, S.M., Dantzic, S., 1989. Eating behavior of women with bulimia. Archives of General Psychiatry 46, 54]58. World Health Organization, 1987. Composite International Diagnostic Interview. WHO, Geneva.
Circulating leptin in patients with anorexia nervosa, bulimia nervosa or binge-eating disorder: relationship to body weight, eating patterns, psychopathology and endocrine changes Palmiero MonteleoneU , Antonio Di Lieto, Alfonso Tortorella, Nicola Longobardi, Mario Maj Institute of Psychiatry, Uni¨ ersity of Naples SUN, Largo Madonna delle Grazie, 80138 Naples, Italy Received 27 September 1999; received in revised form 21 February 2000; accepted 28 February 2000
Abstract A decreased production of leptin has been reported in women with anorexia nervosa ŽAN. and has been attributed merely to the patients’ reduced body fat mass. The extent to which eating patterns, purging behaviors, psychopathology and endocrine changes may contribute to the genesis of leptin alterations has not been deeply investigated. Therefore, we measured plasma levels of leptin, glucose and other hormones in three groups of eating disorder patients with different body weight ŽBW., eating patterns and purging behaviors. Sixty-seven women, 21 with AN, 32 with bulimia nervosa ŽBN., 14 with binge-eating disorder ŽBED. and 25 healthy females volunteered for the study. We found that circulating leptin was significantly reduced in AN and BN patients, but significantly enhanced in women with BED. In anorexics, plasma glucose was decreased, whereas plasma cortisol was enhanced; blood concentrations of 17b-estradiol and prolactin ŽPRL. were reduced in both AN, BN and BED patients. In all subject groups, a strong positive correlation emerged between plasma levels of leptin and the subjects’ BW or body mass index, but not between leptin and psychopathological measures, plasma glucose, cortisol, PRL and 17b-estradiol. Since leptin was reduced in both underweight anorexics and normal weight bulimics, but increased in overweight BED women, who compulsively binge without engaging in compensatory behaviors, we suggest that factors other than BW may play a role in the determination of leptin changes in eating disorders. Q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Anorexia nervosa; Binge-eating disorder; Bulimia nervosa; Leptin
U
Corresponding author. Tel.: q39-081-566-6517; fax: q39-081-566-6523. E-mail address: [email protected] ŽP. Monteleone. 0165-1781r00r$ - see front matter Q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 5 - 1 7 8 1 Ž 0 0 . 0 0 1 4 4 - X
122
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129
1. Introduction
Leptin, the recently discovered adipocyte hormone, is thought to be an important component of the physiological system controlling energy homeostasis. When injected in the experimental animal, leptin behaves as a satiety factor decreasing food intake, increasing caloric expenditure and reducing body weight ŽBW. ŽMantzoros and Moschos, 1998.. Therefore, leptin contributes to BW regulation through modulating feeding behavior and energy expenditure. Moreover, leptin affects several neuroendocrine axes ŽAhima et al., 1996. and is currently thought to be an important factor triggering the onset of puberty and regulating the menstrual cyclicity in humans ŽApter, 1997.. Anorexia nervosa ŽAN. and bulimia nervosa ŽBN. are psychiatric disorders characterized by abnormal eating patterns and purging behaviors, aiming to reduce BW in AN and to prevent its increase in BN, because of the patients’ pathological fear of becoming fat. Malnutrition-induced metabolic and hormonal abnormalities occur in both syndromes and amenorrhea is a key diagnostic symptom of AN, while oligomenorrhea or amenorrhea may be present in BN. Given this background, the investigation of leptin production is of obvious interest in patients with eating disorders. It has been recently shown that, in underweight anorexic patients, leptin plasma levels are severely reduced and correlate with decreased body mass index ŽBMI. and body fat content ŽFerron et al., 1997; Hebebrand et al., 1997; Mantzoros et al., 1997; Kopp et al., 1998; Mathiak et al., 1999.. During the recovery of BW, leptin concentrations progressively re-increase to reach even higher than normal blood values ŽHebebrand et al., 1997.. Therefore, the reduction of body fat mass and BW has been maintained to be the major determinant of the lowered leptin secretion in AN. Some studies have assessed leptin production in BN ŽFerron et al., 1997; Kopp et al., 1997, 1998; Mathiak et al., 1999; Nakai et al., 1999.. However, most of them lack healthy controls
ŽKopp et al., 1997, 1998; Mathiak et al., 1999., one includes patients under different stages of treatment ŽFerron et al., 1997., and one, although including normal controls and untreated patients, groups subjects on the basis of their eating behavior score and not by diagnostic categories ŽNakai et al., 1999.. Hence, the results they provide cannot be conclusive. Besides AN and BN, DSM-IV includes in appendix B Žcriteria sets and axes provided for further studies. the category of binge-eating disorder ŽBED., which is characterized by bingeing episodes and eating-related psychopathology, as in BN, without compensatory behaviors. Therefore, this disorder offers the opportunity to study binge-related alterations without the confounding effects of compensatory behaviors. To the best of our knowledge, only one study has assessed leptin production in obese patients with BED ŽKarhunen et al., 1997.. Therefore, at present, investigations of leptin production in patients with eating disorders are warranted. In the present study, we measured plasma levels of leptin and other hormones in healthy women and in three groups of patients, consisting of women with AN, women with BN and women with BED, and assessed possible relationships between this hormone and demographic, clinical, psychopathological and biochemical variables of eating disorder patients.
2. Methods 2.1. Subjects A total of 92 women were recruited for the study. They were 67 outpatients attending the Eating Disorder Center of our Institute and 25 healthy controls. According to DSM-IV criteria, 21 patients fulfilled the diagnosis of AN, 32 the diagnosis of BN, and 14 the diagnosis of BED. Diagnostic assessment was made by a trained interviewer using the Composite International Diagnostic Interview ŽCIDI. ŽWHO, 1987.. Of the AN women, 14 were exclusively food restrictors
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129
and seven occasionally binged with a frequency no greater than twice a week. At the time of the study, two anorexics had a concomitant diagnosis of obsessive-compulsive disorder and two of major depression. Patients with BN were all of the purging subtype, with binge episodes always followed by self-induced vomiting; three of these patients also abused laxatives and two more exercised excessively. Three had a positive history of AN, two had a concomitant diagnosis of major depression, and one of borderline personality disorder. All of the AN women and nine bulimics were amenorrheic; six bulimics were oligomenorrheic; the remaining patients had normal regular menses. At the time of the study, all patients had been drug-free for more than 6 weeks. Control women were mentally healthy as assessed by the CIDI and had no positive family history of mental disorders as assessed by the Family History Research Diagnostic Criteria ŽAndreasen et al., 1977.. 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 a normal electrocardiogram. Female controls and patients who were normally menstruating were tested in the follicular phase of their menstrual cycle Ždays 5]10 from menses.. None of the subjects were taking oral contraceptives or had a past history of alcohol or drug abuse.
123
after an overnight fast. Blood was collected in tubes with lithium heparin as anticoagulant. Plasma was separated by centrifugation and stored at y208C. 2.3. Biochemical analyses Plasma prolactin ŽPRL. and 17b-estradiol levels were determined by an immunometric method ŽMAIA clone., using commercial kits purchased from Biochem Immuno System ŽMilan, Italy.. The lower detection limits were 0.4 ngrml for PRL and 5 pgrml for 17b-estradiol. Intra- and interassay coefficients of variation were 5.4% and 8.6% for PRL, and 4.3% and 3.2% for 17b-estradiol. Plasma cortisol concentrations were determined by a double-antibody RIA method, using commercial kits purchased from Biochem Immuno System ŽMilan, Italy.. The lower detection limit was 27 nmolrl. Intra- and inter-assay coefficients of variation were less than 5% and 8%. Plasma leptin values were determined by a sandwich enzyme-linked immunosorbant assay, using a commercial kit purchased from Alexis Biochemicals ŽLaufelfingen, Switzerland.. The sensitivity of the method was 0.2 ngrml; intraand inter-assay coefficients of variation were 6.1% and 8.5%, respectively. Plasma glucose was determined by a commercial enzymatic UV method ŽSigma Diagnostics, St. Louis, MO, USA.. 2.4. Data analysis
2.2. Procedure Subjects gave written informed consent prior to study participation. Both patients and healthy subjects underwent a psychopathological assessment with the Eating Disorder Inventory ŽEDI. ŽGarner et al., 1983. and the Bulimic Investigation Test Edinburgh ŽBITE. ŽHenderson and Freeman, 1987. to evaluate eating-related psychopathology; and the Hamilton Depression Rating Scale ŽHDRS. ŽHamilton, 1960. to measure concomitant depressive symptoms. Each subject underwent a blood sample collection in the morning, between 8.00 and 9.00 h,
As there were significant deviations from normality in the data, non-parametric statistical analyses were used. Where the Kruskal]Wallis analysis of variance ŽANOVA. showed significant differences among the groups, the Mann]Whitney U-test was employed to assess differences between patients and controls. The Spearman rank order correlation test was used to examine the relation between leptin values, on the one hand, and demographic, clinical or hormonal data, on the other. In addition, a multiple regression analysis with a backward stepwise procedure was used to define the biochemical variables most predictive of leptin concentrations.
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129
124
3. Results
lated psychopathology and depression ŽTable 1.. No significant differences were observed in any psychometric measure between anorexics of the restricting vs. the binge-purging type.
3.1. Demographic and clinical data
Kruskal]Wallis ANOVA showed significant inter-group differences in BW Ž H s 55.457, d.f.s 3, P- 0.0001. and BMI Ž H s 60.286, d.f.s 3, P0.0001. and percent of ideal BW Ž H s 60.035, d.f.s 3, P- 0.0001., but not in age, height or duration of the illness ŽTable 1.. With respect to healthy women, AN patients had significantly lower BW Ž z s 5.424, P- 0.0001, Mann]Whitney test., BMI Ž z s 5.711, P- 0.0001. and percent of ideal BW Ž z s 5.777, P - 0.0001., whereas patients with BED had significantly higher BW Ž z s 4.186, P - 0.0001., BMI Ž z s 4.098, P 0.0001. and percent of ideal BW Ž z s 4.274, P0.0001.. No significant difference in these variables was observed between bulimics and healthy women and between anorexics of the restricted subtype and those of the binge-purging subtype. For the bulimic group, the mean bingervomiting frequency was 1.9" 1.3 episodesrday; for the BED group, the binge frequency was 2.7" 0.8 episodesrday. As compared with healthy women, anorexic, bulimic and BED patients had significantly higher scores on standardized assessments of eating-re-
3.2. Biochemical data
Kruskal]Wallis ANOVA showed significant inter-group differences in plasma levels of leptin Ž H s 60.848, d.f.s 3, P- 0.0001., 17b-estradiol Ž H s 22.868, d.f.s 3, P- 0.0001. and PRL Ž H s 35.814, d.f.s 3, P- 0.002., but not in plasma glucose Ž H s 5.596, d.f.s 3, P s 0.1. or cortisol concentrations Ž H s 6.520, d.f.s 3, Ps 0.08.. As compared with healthy women, AN and BN patients exhibited significantly decreased plasma levels of leptin, whereas women with BED had significantly enhanced blood concentration of the hormone ŽFig. 1.. No significant difference was observed in leptin plasma levels between anorexics of the restricting vs. the binge-purging type. Bulimics with amenorrhea or oligomenorrhea had leptin plasma values that were lower than values in the normally menstruating ones Ž3.16" 1.6 ngrml vs. 6.1" 4.8 ngrml., although the difference did not reach statistical significance. Plasma levels of 17b-estradiol and PRL were significantly reduced in all the patient groups
Table 1 Demographic and clinical characteristics of study sample a
Number of subjects Age Žyears. Body weight Žkg. Height Žcm. Body mass index Žkgrm2 . Percent of ideal body weight EDI total score BITE total score HDRS total score Duration of the illness Žyears.
Healthy women
Women with anorexia nervosa
Women with bulimia nervosa
Women with binge-eating disorder
25 25.4" 4.2 56.9" 6.1 161.7" 4.7 21.7" 2.2 101.2" 10.2 24.4" 15.8 5.7" 3.3 1.3" 2.8 Not applicable
21 25.0" 6.1 41.9" 5.6U 162.0" 5.0 16.0" 1.6U 74.4" 7.2U 72.1" 20.3U 19.9" 9.0U 16.9" 6.7U 5.3" 5.1
32 23.2" 3.3 53.6" 8.8 161.8" 7.7 20.5" 2.8 95.7" 21.4 96.1" 26.8U 36.6" 7.6U 12.6" 8.3U 4.9" 2.8
14 26.0" 6.9 82.2" 20.2U 163.5" 5.6 30.9" 7.9U 144.9" 35.8U 95.4" 22.9U 34.0 " 8.5U 11.9" 7.0U 4.1" 3.5
All data are presented as mean " S.D. U P- 0.0001 vs. healthy women ŽMann]Whitney test..
a
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129
125
Fig. 1. Plasma leptin concentrations in healthy women and in women with anorexia nervosa, bulimia nervosa and binge-eating disorder. Data are expressed as mean " S.E.M. U P- 0.005 and UU P- 0.0001 vs. healthy women ŽMann]Whitney test..
ŽTable 2.. No significant difference in these parameters was observed between anorexics of the restricting vs. the binge-purging type. Eumenorrheic bulimics had plasma levels of 17bestradiol which were higher than those of bulimics with menstrual irregularities Ž zs 1.71, P- 0.04. and not significantly different from those of healthy women. PRL plasma levels did not significantly differ between the two bulimic subgroups. Since in the AN group mean baseline plasma cortisol concentrations were higher and plasma glucose was lower than in healthy subjects, a further ANOVA was run to test for possible
differences between AN patients and control women that could have been missed in a fourgroup comparison. This analysis showed a significant difference between anorexic patients and healthy subjects for both cortisol and glucose values Ž H s 6.859, d.f.s 1, P- 0.01 and H s 5.260, d.f.s 1, P - 0.025, respectively.. The Mann]Whitney test confirmed that anorexics had higher plasma cortisol Ž z s 2.602, P- 0.005. and lower plasma glucose Ž z s 2.304, P- 0.02. concentrations than healthy women with no significant differences between the two AN subgroups. In order to evaluate whether differences in estradiol and cortisol levels between groups per-
Table 2 Comparison between healthy women and women with eating disorders for plasma hormone and glucose concentrations a
Plasma prolactin Žngrml. Plasma cortisol Žnmolrl. Plasma 17b-estradiol Žpgrml. Plasma glucose Žmgrdl.
Healthy women Ž n s 15.
Women with anorexia nervosa Ž n s 21.
Women with bulimia nervosa Ž n s 32.
Women with binge-eating disorder Ž n s 14.
19.0" 7.6 304 " 107 72.4" 59.2 93.6" 20.4
7.2" 2.7U 445 " 171UU 18.7" 16.1U 82.7" 11.5UUU
9.7" 6.5U 370 " 238 42.0" 40.5UU 84.2" 18.4
9.0" 3.9U 350 " 120 34.0" 8.5UUU 90.0" 15.3
All data are presented as mean " S.D. U P- 0.0001; UU P- 0.005; UUU P- 0.02 vs. healthy women ŽMann]Whitney test..
a
126
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129
sisted after adjustment for leptin plasma levels, an analysis of covariance was performed with leptin as the dependent variable. The results of this analysis showed that neither cortisol Ž F s 0.01, d.f.s 1, 43, Ps NS. nor 17b-estradiol Ž F s 0.44, d.f.s 1, 87, Ps NS. was significantly associated with leptin. 3.3. Correlations Leptin plasma levels were positively correlated to both BW and BMI in AN Žr s 0.44, P- 0.02; r s 0.54, P- 0.005, respectively., BN Žr s 0.31, P- 0.04; r s 0.36, P- 0.02. and BED patients Žr s 0.78, P- 0.0003; r s 0.81, P- 0.0001. as well as in healthy women Žr s 0.59, P- 0.0008; r s 0.62, P- 0.0004. ŽFig. 2.. The multiple regression analysis showed that glucose, 17b-estradiol, PRL and cortisol were not
significant determinants of leptin concentrations in our sample.
4. Discussion This is the first study to explore leptin production across the eating-disorder spectrum. We found that circulating morning leptin was significantly reduced in untreated patients with AN and in those with BN, whereas it was significantly increased in women with BED, compared with age-matched healthy females. Consistent with our findings, decreased blood levels of leptin in AN patients have been reported by other authors ŽFerron et al., 1997; Hebebrand et al., 1997; Mantzoros et al., 1997; Kopp et al., 1998; Mathiak et al., 1999., who suggested that the lowered production of the adipocyte hormone
Fig. 2. Relationships between plasma leptin concentrations and body mass index ŽBMI. in healthy women and in patients with anorexia nervosa, bulimia nervosa or binge-eating disorder ŽBED..
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129
merely reflects the reduced amount of body fat mass, which typically occurs in AN. Previous studies exploring leptin production in BN are not directly comparable with our study because of the lack of normal controls ŽKopp et al., 1997, 1998; Mathiak et al., 1999., the inclusion of patients at different stages of nutritional and behavioral treatment ŽFerron et al., 1997. or the grouping of patients on the basis of their eating behavior score and not by diagnostic categories ŽNakai et al., 1999.. In agreement with previous reports ŽFerron et al., 1997; Hebebrand et al., 1997; Mantzoros et al., 1997; Kopp et al., 1998; Mathiak et al., 1999., our results show an unambiguous positive correlation between circulating leptin and BW or BMI in all the subject groups. However, our results also suggest that the reduction in BW is not the only determinant of the lowered secretion of leptin in eating-disorder patients. Indeed, bulimic women, who have abnormal eating behavior, like the anorexic ones, did show a clear-cut decrease in circulating leptin in spite of the maintenance of a normal BW. This observation strongly indicates that, in addition to BW, other factors play a role in regulating leptin production in eating disorders. Restrained eating has recently been shown to be associated with low leptin levels in young women Žvon Prittwitz et al., 1997.. Restrained eating occurs in both patients with AN and those with BN, although in the latter it is not associated with decreased BW. Restrained eaters restrict their energy intake to avoid weight gain and bulimics, despite the large caloric ingestion during binge episodes, regurgitate a relatively large amount of calories by vomiting and engage in prolonged starvation in order to reduce their daily caloric ingestion. A caloric intake below the normal requirements of the respective individual, even with no loss of fat mass, may have an impact on leptin synthesis. Therefore, one would suggest that, besides BW, the eating pattern is important in the modulation of leptin secretion in humans. Although in our study we did not measure the eating behavior of our patients by ad hoc psychometric scales, it is worth mentioning that plasma levels of leptin have been shown to significantly
127
correlate with the eating behavior score in a cohort of patients with AN or BN ŽNakai et al., 1999.. Furthermore, in the present study, patients with BED, who compulsively binge, but do not fast or engage in compensatory behaviors, exhibited an increased concentration of plasma leptin as one would normally expect according to their enhanced BW. Therefore, it seems possible to conclude that low plasma leptin may be a sensitive indicator not only of the overall nutritional status, but also of metabolic changes occurring with restrained eating and purging behaviors. Since leptin behaves as a satiety factor, it is possible to speculate that its reduced levels in both AN and BN could contribute to the increased hunger sensation of the anorexic women ŽOwen et al., 1985. and to the binge-eating behavior of bulimics. Indeed, human laboratory studies suggested that bulimic patients have diminished satiety responses ŽWalsh et al., 1989. that, on the basis of our results, could be mediated by a decreased production of leptin. This may lead to binge eating. In women with BED, who exhibit increased levels of leptin, an impaired sensitivity to the enhanced circulating leptin may be involved in the pathogenesis andror maintenance of the bingeing behavior. In most cases, BED patients are truly obese, and a reduced sensitivity to the abnormally high circulating leptin levels has been hypothesized in human obesity ŽMantzoros and Moschos, 1998.. Future studies assessing the relationship between plasma leptin and satiety or hunger sensations may help to confirm or disconfirm this hypothesis. A further observation of our study is that, in the BN group, patients with amenorrhea or oligomenorrhea exhibited lower circulating leptin levels than those with regular menses, although this difference did not reach statistical significance. This finding confirms previous suggestions that reduced leptin may be an important determinant of the menstrual abnormalities occurring in patients with eating disorders ŽKopp et al., 1997., even if, due to the relatively small number of subjects in our subgroups of bulimic women, a type-II statistical error cannot be ruled out. According to previous findings by both other authors and our group ŽSchreiber et al., 1991;
128
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129
Monteleone et al., 1998a,b, 1999; Stoving et al., 1999., we observed decreased blood levels of glucose and increased concentrations of cortisol in AN women, and reduced plasma values of 17bestradiol and PRL in all the patient groups. It has been suggested that blood glucose may influence leptin secretion likely by inducing insulin production ŽBoden et al., 1996.. Insulin, in fact, is believed to modulate leptin synthesis by both enhancing leptin mRNA expression and exerting a trophic effect on the adipocytes ŽMantzoros and Moschos, 1998.. Similarly, estrogens and glucocorticoids have been reported to increase leptin production in humans ŽElbers et al., 1997; Newcomer et al., 1998., and leptin, in turn, has been shown to regulate the circadian oscillations of both luteinizing hormone and estradiol in normal women ŽLicinio et al., 1998. and to inhibit the production of cortisol ŽBornstein et al., 1997., which suggests the existence of feedback loops between leptin and both ovarian and adrenal steroids. Therefore, one could speculate that, in our eating-disorder patients, the changes in plasma levels of leptin, glucose, 17b-estradiol and cortisol may be mutually linked. However, on the one hand, we found that leptin did not affect the differences in 17b-estradiol and cortisol levels among the groups, and, on the other hand, neither the hormonal parameters nor glucose plasma levels proved to be a significant determinant of leptin plasma concentrations. Therefore, it seems possible to exclude that glucose and endocrine abnormalities are linked to the altered leptin secretion in eating disorder patients. Some limitations of the present study need to be discussed. First of all, it must be noted that body fat correlates with leptin levels to a higher degree than does BW, mainly in patients with eating disorders ŽMathiak et al., 1999.. Hence, a shortcoming of the present study is that we did not measure body fat, which may have differed between groups independently of BMI. However, within the same gender and age range, it has been shown that BMI may represent average body fat estimates ŽHeitmann, 1990.. A further limitation of our study is that our findings refer to a single morning measurement of blood hormones in spite of the well-recognized
circadian rhythmicity in the secretion of these substances. However, although we cannot exclude that phase shifts in the circadian rhythms may have affected our results, it is of note that the present findings are in line with previously reported values of morning plasma leptin, cortisol and PRL in eating-disorder patients. Finally, the measurement of blood levels of insulin, insulinlike growth factor I and triiodothyronine would have been of interest, since these hormones have been linked to leptin production in eating-disorder patients ŽGrinspoon et al., 1996; Nakai et al., 1999.. In conclusion, circulating leptin is reduced in untreated underweight anorexic patients and normal weight bulimics, whereas it is increased in overweight BED patients. BW, menstrual status, eating patterns and purging behaviors, but not psychopathology or other metabolic and endocrine abnormalities, seem to be the major determinants of leptin changes in eating disorders. References Ahima, R.S., Prabakaran, D., Mantzoros, C., Qu, D.Q., Lowell, B., Maratosflie, E., Flier, J.S., 1996. Role of leptin in the neuroendocrine response to fasting. Nature 382, 250]252. Andreasen, N.C., Endicott, J., Spitzer, R.L., Winokur, G., 1977. The family history method using diagnostic criteria: reliability and validity. Archives of General Psychiatry 34, 1229]1235. Apter, D., 1997. Leptin in puberty. Clinical Endocrinology 47, 175]176. Boden, G., Chen, X., Mozzoli, M., Ryan, I., 1996. Effect of fasting on serum leptin in normal human subjects. Journal of Clinical Endocrinology and Metabolism 81, 3419]3423. Bornstein, S.R., Uhlmann, K., Haidan, A., Ehrhart-Bornstein, M., Scherbaum, W.A., 1997. Evidence for a novel peripheral action of leptin as a metabolic signal to the adrenal gland: leptin inhibits cortisol release directly. Diabetes 46, 1235]1238. Elbers, J.M.H., Asscheman, H., Seidell, J.C., Frolich, M., Meinders, A.E., Gooren, L.J.C., 1997. Reversal of the sex difference in serum leptin levels upon cross-sex hormone administration in transsexuals. Journal of Clinical Endocrinology and Metabolism 82, 3267]3270. Ferron, F., Considine, R.V., Peino, R., Lado, I.G., Dieguez, C., Casanueva, F.F., 1997. Serum leptin concentrations in patients with anorexia nervosa, bulimia nervosa and nonspecific eating disorders correlate with the body mass index but are independent of the respective disease. Clinical Endocrinology 46, 289]293.
P. Monteleone et al. r Psychiatry Research 94 (2000) 121]129 Garner, D.L., Olmstead, M.P., Polivy, J., 1983. Development and validation of a multidimensional eating disorders inventory for anorexia nervosa and bulimia. International Journal of Eating Disorders 2, 15]35. Grinspoon, S., Gulick, T., Askari, H., Landt, M., Lee, K., Anderson, E., Ma, Z.M., Vignati, L., Bowsher, R., Herzog, D., Klibanski, A., 1996. Serum leptin levels in women with anorexia nervosa. Journal of Clinical Endocrinology and Metabolism 81, 3861]3863. Hamilton, M., 1960. A rating scale for depression. Journal of Neurology, Neurosurgery and Psychiatry 23, 56]62. Hebebrand, J., Blum, W.F., Barth, N., Coners, H., Englaro, P., Juul, A., Ziegler, A., Warnke, A., Rascher, W., Remschmidt, H., 1997. Leptin levels in patients with anorexia nervosa are reduced in the acute stage and elevated upon short-term weight restoration. Molecular Psychiatry 2, 330]334. Heitmann, B.L., 1990. Evaluation of body fat estimated from body mass index, skinfolds and impedance. A comparative study. European Journal of Clinical Nutrition 44, 831]837. Henderson, M., Freeman, C.P.L., 1987. A self-rating scale for bulimia: the BITE. British Journal of Psychiatry 150, 18]24. Karhunen, L., Haffner, S., Lappalainen, R., Turpeinen, A., Miettinen, H., Uusitupa, M., 1997. Serum leptin and shortterm regulation of eating in obese women. Clinical Sciences 92, 573]578. Kopp, W., Blum, W.F., von Prittwitz, S., Ziegler, A., Lubbert, H., Emons, G., Herzog, W., Herpertz, S., Deter, H-C., Remschmidt, H., Hebebrand, J., 1997. Low leptin levels predict amenorrhea in underweight and eating disordered females. Molecular Psychiatry 2, 335]340. Kopp, W., Blum, W.F., Ziegler, A., Mathiak, K., Lubbert, H., Herpertz, S., Deter, H.C., Hebebrand, J., 1998. Serum leptin and body weight in females with anorexia and bulimia nervosa. Hormone and Metabolic Research 30, 272]275. Licinio, J., Negrao, A.B., Mantzoros, C., Kaklamani, V., Wong, M.-L., Bongiorno, P.B., Mulla, A., Cearnal, L., Veldhuis, J.D., Flier, J.S., McCann, S.M., Gold, P.W., 1998. Synchronicity of frequently sampled, 24-h concentrations of circulating leptin, luteinizing hormone, and estradiol in healthy women. Proceedings of the National Academy of Sciences USA 95, 2541]2546. Mantzoros, C.S., Moschos, S.J., 1998. Leptin: in search of roles in human physiology and pathophysiology. Clinical Endocrinology 49, 551]567. Mantzoros, C., Flier, J.S., Lesem, M.D., Brewerton, T.D., Jimerson, D.C., 1997. Cerebrospinal fluid leptin in anorexia
129
nervosa: correlation with nutritional status and potential role in resistance to weight gain. Journal of Clinical Endocrinology and Metabolism 82, 1845]1851. Mathiak, K., Gowin, W., Hebebrand, J., Ziegler, A., Blum, W.F., Felsenberg, D., Lubbert, H., Kopp, W., 1999. Serum leptin levels, body fat deposition, and weight in females with anorexia or bulimia nervosa. Hormone and Metabolic Research 31, 274]277. 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. Psychological Medicine 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. British Journal of Psychiatry 172, 438]442. Monteleone, P., Brambilla, F., Bortolotti, F., Maj, M., 1999. Serotonergic dysfunction across the eating disorders: relationship to eating behaviour, purging behaviour, nutritional status and general psychopathology. Submitted. Nakai, Y., Hamagaki, S., Kato, S., Seino, Y., Takagi, R., Kurimoto, F., 1999. Leptin in women with eating disorders. Metabolism 48, 217]220. Newcomer, J.W., Selke, G., Melson, A.K., Gross, J., Vogler, G.P., Dagogo-Jack, S., 1998. Dose-dependent cortisol-induced increases in plasma leptin concentration in healthy humans. Archives of General Psychiatry 55, 995]1000. Owen, W.P., Kalmi, K.A., Gibbs, J., Smith, G.P., 1985. Satiety responses in eating disorders. Journal of Psychiatric Research 19, 279]284. Schreiber, W., Schweiger, U., Werner, D., Brunner, G., Tuschl, R.J., Laessle, R.G., Krieg, J.C., Fichter, M.M., Pirke, K.M., 1991. Circadian pattern of large neutral amino acids, glucose, insulin, and food intake in anorexia nervosa and bulimia nervosa. Metabolism 40, 503]507. Stoving, R.K., Hangaard, J., Hansen-Nord, M., Hagen, C., 1999. A review of endocrine changes in anorexia nervosa. Journal of Psychiatric Research 33, 139]152. von Prittwitz, S., Blum, W.F., Ziegler, A., Scharmann, S., Remschmidt, H., Hebebrand, J., 1997. Restrained eating is associated with low leptin levels in underweight females. Molecular Psychiatry 2, 420]422. Walsh, T.B., Kissileff, H.R., Cassidy, S.M., Dantzic, S., 1989. Eating behavior of women with bulimia. Archives of General Psychiatry 46, 54]58. World Health Organization, 1987. Composite International Diagnostic Interview. WHO, Geneva.