Persistent amenorrhoea in weight-recovered anorexics: psychological and biological aspects

Psychiatry Research 118(2003) 249–257

0165-1781/03/$ - see front matter� 2003 Elsevier Science Ireland Ltd. All rights reserved.doi:10.1016/S0165-1781Ž03.00074-X

Persistent amenorrhoea in weight-recovered anorexics:psychological and biological aspects

Francesca Brambilla *, Palmiero Monteleone , Francesca Bortolotti ,a, a a

Riccardo Dalle Grave , Patrizia Todisco , Angela Favaro , Paolo Santonastaso ,b b c c

Carla Ramacciotti , Roberto Paoli , Mario Majd d a

Istituto di Clinica Psichiatrica, Universita, SUN, Naples, Italya `Villa Garda, Garda, Verona, Italyb

Istituto di Clinica Psichiatrica, Universita, Padova, Italyc Ìstituto di Clinica Psichiatrica, Universita, Pisa, Italyd `

Received 24 April 2002; received in revised form 17 December 2002; accepted 19 February 2003

Abstract

Demographic, psychopathological and hormonal parameters of 22 women with previous anorexia nervosa(AN)presently recovered, in a state of stabilized nutritional normalization for 3 months to 2 years but with persistentamenorrhoea, and of 20 psychophysically healthy age- and sex-matched normally menstruating controls were studied.Body mass index(BMI) values did not differ in patients and controls. Psychological examination, monitored byEating Disorder Inventory 1, Bulimic Investigation Test Edinburgh, Yale-Brown-Cornell Eating Disorder Scale, andTridimensional Personality Questionnaire rating scales, showed the persistence of some of the psychopathologicalsymptoms of AN. Hormonal examinations included basal plasma concentrations of follicle stimulating hormone,luteotropic hormone, estrogens(E), progesterone, thyrotropic hormone, FT , FT(immunoradiometric assays), leptin3 4

(LEP) (enzymatic-linked-immunosorbent assay) and 24 h urinary free cortisol(immunoradiometric assay). Hormonevalues were the same in patients and controls, except for E and LEP levels, which were significantly lower in patientsthan in controls. The concentrations of the two hormones were not correlated with the BMI of the patients, but LEPvalues were correlated negatively with the difference between the present BMI and the preanorexic one. The valuesof both hormones correlated negatively with some of the psychopathological aspects typical of AN, in particular withhigh ‘body dissatisfaction’, ‘ineffectiveness’, and ‘interpersonal distrust’ and with low ‘interoceptive awareness’.� 2003 Elsevier Science Ireland Ltd. All rights reserved.

Keywords: Anorexia nervosa; Amenorrhoea; Leptin; Estradiol; Eating Disorder Inventory

*Centro di Psiconeuroendocrinologia, Piazza Grandi 3, Milan 20129, Italy. Tel.:q39-368-3017420; fax:q39-02-70122889.E-mail address: [email protected](F. Brambilla).

250 F. Brambilla et al. / Psychiatry Research 118 (2003) 249–257

1. Introduction

Amoenorrhea is a mandatory symptom for thediagnosis of anorexia nervosa(AN) according tothe DSM-IV criteria(American Psychiatric Asso-ciation, 1994). It develops at the beginning of theeating disorder and remits with recovery from it,having been linked pathogenetically to the extremeweight loss and the micro–macronutrient deficien-cy that characterize the disease. However, it hasalso been reported that amenorrhoea may start inAN before weight loss is really significant andmay persist long after resumption of proper eatingand weight recovery, suggesting that the nutritionalcomponent of the disease may not be the onlypathogenetic factor for the appearance and main-tenance of the menstrual disorder(Dally and Sar-gant, 1966; Bell et al., 1966; Russell, 1972; Valckeet al., 1974; Frisch and McArthur, 1974; Warrenet al., 1975; Beumont et al., 1976; Wackeling etal., 1976, 1977; Katz et al., 1978; Falk and Halmi,1982; Lecomte et al., 1984; Buvat et al., 1984;Nakamura et al., 1985; Meyer et al., 1986a,b;Kohmura et al., 1986; Treasure et al., 1988; Jam-rozik et al., 1995). It has been proposed that long-lasting amenorrhoea that persists after weightrecovery is due to AN-related alterations of theneuroendocrine axis that modulate the occurrenceand the rhythmicity of menses. Data in the litera-ture for weight-recovered amenorrhoeic anorexicsreport impairments of the hypothalamo-pituitary-gonadal (HPG) axis, with lower than normalluteotropic hormone (LH) concentrations,increased variability of LH levels, reduced LHpulse numbers, absence of a mature LH secretorypattern, increased immunoreactive LH half-life,and lower than normal plasma estradiol(E) levels.Instead, no secretory alterations were observed forfollicle stimulating hormone(FSH), prolactin, cor-tisol (CORT), thyrotropic hormone(TSH), thyrox-ine (T ) and triiodothyronine(T ), all of which4 3

modulate the regular occurrence of menses(Katzet al., 1978; Lecomte et al., 1984; Meyer et al.,1986a,b; Jamrozik et al., 1995). The data in theliterature, however, do not explain why somepatients with substantial eating normalization andweight recovery are still amenorrhoeic and havethe above-mentioned neuroendocrine alterations

while other nutritionally recovered anorexicsresume regular menses and normal neuroendocrinefunction. It has been mentioned, very generically,that the amenorrhoea and the related neuroendo-crine alterations may result from impairments ofneurotransmitters that regulate HPG secretion, butno data for this have been reported up to now.Studies of the psychological aspects of recov-

ered anorexics with persistent amenorrhoea haveshown that some anorexia-related symptoms stillremain in these subjects, including disturbed per-ceptions of body and of inner stimuli, lack of asense of well-being, impaired sexual behavior(Meyer et al., 1986a,b), thin body ideal, fear ofbecoming fat, a compulsive eating, selective appe-tite, desire to control the environment, and obses-sive-compulsive behavior(Falk and Halmi, 1982).These psychopathological aspects might predictrelapsing of AN. No explanations, however, wereoffered for the biological mechanisms that mightlink the psychopathology to the neuroendocrinedysfunction.Since the persistence of amenorrhoea and the

related biological and psychological alterationsmight be negative predictors for a favourable andstable outcome of AN, we decided to investigatethe demographic, psychopathological and neuroen-docrine aspects of weight-recovered, properly eat-ing, amenorrhoeic patients, and the relationshipsbetween these three parameters, alterations ofwhich might possibly be etiopathogenetic for thepersistence of the mental disorder.

2. Material and methods

Twenty-two female Caucasian outpatients withprevious histories of AN and presently consideredrecovered according to DSM-IV(American Psy-chiatric Association, 1994), 16 of the restrictedtype (AN-R) and six of the bingeing-purging type(AN-BP), entered the study. They were recruitedin the University Psychiatric Institutes of Naples,Padua and Pisa and in the hospital ‘Villa Garda’in Garda-Verona, Italy. Consensus diagnosesaccording to DSM-IV criteria were obtained bysenior psychiatrists who assessed the patients inde-pendently by clinical interviews and by the Struc-tured Clinical Interview for DSM-IV (SCID;

251F. Brambilla et al. / Psychiatry Research 118 (2003) 249–257

Table 1Demographic data of anorexics and controls

Anorexics Controls(ns22) (ns20)

Age (years) 23.4"4.3 23.3"4.4Age of onset of AN(years) 18.4"4.2Duration of AN (years) 4.7"4.3Onset of amenorrhea(years) 18.0"1.0Duration of amenorrhoea(years) 4.5"0.9BMI before AN 20.7"2.2BMI nadir 14.5"2.1BMI after AN 19.3"1.1 20.4"1.2

Means"S.D.; AN, anorexia nervosa; BMI, body massindex; BMI nadir, minimum BMI reached during the disease.

Spitzer, 1994). Twenty physically healthy Cauca-sian females, recruited from the hospital staff,formed the age-matched control group. They werealso interviewed with the SCID to exclude presentand lifetime Axis I or II disorders, and especiallydisorders of eating behavior. Patients and controlsgave informed consent to participate in the study,after an extensive description of its protocol.The demographic data for patients and controls

are reported in Table 1. Anorexics were consideredclinically recovered since they had reached presentnormal weights and body mass index(BMI) supe-rior to 18.5 (as defined in DSM-IV), had beeneating properly in terms of quantity and selectionof food, and bingeing-purging had stopped from 3months to 2 years(mean"S.D.s1.8"1.01)before our study. Present eating habits were con-trolled by a diary kept in the 7 days prior to ourstudy. Food consumption was considered normalwhen the patients were consuming between 1900and 2300 caloriesyday (mean"S.D.s1927.6"467,9), properly subdivided in the different micro–macronutrients(52% carbohydrates, 28% lipids,20% proteins). Controls were consuming the sameamount of calories. Both patients and controlswere not exercising at the moment of our studyand at least in the previous 3 months. Amenorrhoeahad started from 2 weeks to 3 months after thebeginning of AN and persisted until the presentobservations. The controls were examined duringthe early follicular phase of the menstrual cycle toexclude cycle-related sex hormone fluctuations.

Before our study, some of the patients had beentreated with selective serotonin reuptake inhibitorsincluding paroxetine, fluoxetine and citalopram,which were stopped at least 6 months before thestart of our investigation.Criteria for exclusion from our study for patients

and controls were organic diseases, immunopa-thies, allergopathies, endocrinopathies(apart fromthose related to the hypotalamo-pituitary-gonadalaxis in AN), alcohol or drug abuse, use of contra-ceptive drugs, brain organic disorders, previouscerebral trauma, and other Axis I and II patholo-gies(with the exclusion of AN for the patients).The psychopathological aspects of the patients

were assessed with the eating disorder inventory 1(E.D.I.-1; Garner et al., 1983), the Bulimic Inves-tigation Test Edinburgh(B.I.T.E.; Henderson andFreeman, 1987), the Yale-Brown-Cornell EatingDisorder Scale (Y-BOCS-ED; Mazure et al.,1994), and the Tridimensional Personality Ques-tionnaire(TPQ; Cloninger et al., 1994).

The neuroendocrine examination includedassays of basal plasma concentrations of FSH andLH, E, progesterone(PROG), TSH, FT , FT , and3 4

leptin (LEP), and of the urinary 24 h total freeCORT values. An ultrasonographic scan wasobtained in each patient to exclude non-anorexic-related ovarian pathology.Patients and controls arrived at the day hospital

of our institutes at 08.30 h for blood collection forhormone assays, after fasting in the previous 12 h.A cannula was inserted at 08.30 h into an antecu-bital vein, kept patent by saline infusion. At 09.00h, edetic acid anticoagulated blood was drawn,immediately centrifuged and the plasma frozen aty20 8C until assayed. The day before blooddrawing patients and controls collected 24 h urinesfor the free CORT assay, which were delivered tothe day hospital at the time of the blood drawing.The hormone assays were done by immunora-

diometric methods for FSH, LH, E, PROG, FT ,4

FT , and TSH using the commercial kits of Bioch-3

em Immunosystem(Milano, Italy), and with asand- wich enzyme-linked immunosorbent assayfor LEP, using commercial kits from Alexis Bioch-emicals (Laufelfingen, Switzerland). Interassaycoefficients of variations were-11% for LH, -8% for FSH, 3.2% for E, 12.5% for PROG, 8%


Table 2Psychological aspects of anorexics and controls

Patients Controls P(ns22) (ns20)

E.D.I.-1 total 54.0"26.5 25.5"12.7 -0.0002Drive for thinness 7.5"7.3 2.1"2.6 -0.01Bulimia 7.0"7.2 2.2"2.4 NSBody dissatisfaction 9.2"6.5 0.7"1.4 -0.00001Ineffectiveness 6.4"5.2 5.2"4.2 NSPerfectionism 5.3"3.5 2.7"3.8 -0.005Interpersonal distrust 6.2"4.5 5.8"2.9 NSInteroceptive awareness 7.5"6.7 5.8"2.9 -0.03Maturity fear 7.8"3.5 3.4"3.0 NS

B.I.T.E. 15.0"12.2 7.0"5.3 -0.02Y-BOCS-E.D. 6.5"6.7 0 -0.0002HA 19.1"5.6 15.9"2.8 -0.02SS 13.8"4.1 17.2"5.2 -0.04Reward dependance 18.5"5.8 18.5"3.2 NS

Means"S.D.; E.D.I.-1, Eating Disorder Inventory; B.I.T.E.,Bulimic Investigation Test Edinburgh; Y-BOCS-E.D., Yale-Brown Obsessive Compulsive Scale for Eating Disorder.

Table 3Hormonal parameters of anorexics and controls

Patients Controls P(ns22) (ns20)

FSH (mIUyml) 6.1"2.6 5.0"1.9 NSLH (mIUyml) 2.6"2.2 3.6"1.3 NSE (pgyml) 46.1"31.9 82.7"61.4 -0.03PROG(ngyml) 0.6"0.5 1.3"0.7 -0.002CORT (ngyml) 125.8"58.0 94.9"47.6 NSTSH (mIUyml) 1.6"0.9 1.7"0.7 NSFT (pmolyl)3 3.0"3.1 2.1"0.9 NSFT (pmolyl)4 1.2"0.4 1.3"0.2 NSLEP (ngyml) 3.8"3.1 10.1"6.3 -0.0003

Means"S.D.; FSH, follicle stimulating hormone; LH, luteo-tropic hormone; E, 17b-estradiol; PROG, progesterone; CORT,cortisol; TSH, thyro-tropic hormone; FT , free triiodothyro-3

nine; FT , free thyroxine; LEP, leptin.4

for CORT, 4.6% for TSH, 4.6% for T , 8% for3

T , and 5% for LEP. Intraassay coefficients of4

variation were-7 for LH, -4.5% for FSH, 4.3%for E, 7.3% for PROG, 5% for CORT, 3.4% forTSH, 6% for T , 5.7% for T , and 6.1% for LEP.4 3

The statistical significance of the data wasevaluated by nonparametric Mann–Whitney anal-ysis comparing data of patients to those of controlsand by Student’st-test for paired data to comparevalues from the same subjects at different timepoints. The Spearman rank correlation coefficientwas applied to analyze the relationships betweenhormonal concentrations, demographic data andscores of the psychometric rating scales. To dealwith the problem of statistical significance inmultiple correlation tests, a stepwise multipleregression analysis was performed in order toassess the weight of each independent variable onboth LEP and estrogen variability.

3. Results

Since no significant differences were observedin the demographic, psychopathological and hor-monal data for AN-R and AN-BP patients, theywere analyzed and reported together. BMI valuesof patients at the time of our study and those of

controls were not statistically different. Instead,values of patients before the beginning of AN andat the time of our investigation were significantlydifferent (Ps0.005,ts2.94).The data from the psychological parameters,

collected with the rating scales, are reported inTable 2. E.D.I.-1 values were significantly higherfor patients than controls(P-0.0002,zsy2.84).Among the subtypes of E.D.I.-1 scores, values of‘drive for thinness’, ‘body dissatisfaction’, ‘perfec-tionism’ and ‘interoceptive awareness’ were sig-nificantly higher for patients than controls(respectively, P-0.01, zsy2.38; P-0.00001,zsy5.24; P-0.005, zsy2.79; P-0.03, zsy2.1), while there were no significant differencesfor the other subtypes. B.I.T.E. values were signif-icantly higher for patients than controls(P-0.02,zsy1.18). Y-BOCS-ED values were higher forpatients than controls(Ps0.0002, zsy3.6).Among the temperament traits, scores of harmavoidance (HA) were significantly higher forpatients than controls(P-0.02,zsy2.269), thoseof sensation seeking(SS) were significantly lower(P-0.04,zsy2.06), and there was no differencebetween the two groups for reward dependence(RD).The hormonal data are reported in Table 3. No

statistically significant differences between patientsand controls were observed for concentrations ofFSH, LH, CORT, TSH, FT and FT , while those3 4

of E and LEP were significantly lower in patients


than in controls(EsP-0.03, zsy2.07; LEPsP-0.00003,zsy4.16). PROG levels were alsosignificantly lower in patients than in controls(PROGsP-0.002, zsy3.08), but in patientsthey could still be included among values typicalfor the early follicular phase of the cycle. Therewere no correlations between concentrations ofeach hormone and demographic data(age, BMI)or psychological parameters in patients or controls.Only values of LEP and E concentrations showedsome intriguing correlations.LEP levels were not correlated with body weight

or BMI levels but were correlated negatively withthe difference between the BMI values before thebeginning of AN and those after weight recovery(P-0.04, rsy0.44). No correlation wasobserved between LEP levels and duration ofweight restoration. Instead, LEP levels correlatednegatively with E.D.I.-1 total scores(P-0.005,rsy0.42), and in detail with the subitem scores‘drive for thinness’(P-0.03, rsy0.32), ‘bodydissatisfaction’(P-0.0008,rsy0.57), ‘maturityfear’ (P-0.007, rsy0.40) and ‘bulimia’ (P-0.03, rsy0.32). Because of these statisticallysignificant multiple correlations, a stepwise multi-ple regression analysis was applied to assess theinfluence of each EDI psychopathological dimen-sion and of changes in body weight or plasmaLEP variability. This analysis showed that only‘body dissatisfaction’ scores were found to signif-icantly affect LEP values(F s7.73,Ps0.008,1.40

Rs0.16). There were no correlations with B.I.T.E.,Y-BOCS-ED or TPQ scores.E values were not correlated with any demo-

graphic (in particular BMI, difference betweenBMI or body weight before starting AN and at thetime of our study, and duration of weight restora-tion) or hormonal data. They were correlatednegatively with total E.D.I.-1 scores(P-0.05,rsy0.30), and in detail with ‘drive for thinness(P-0.01,rsy0.36), ‘body dissatisfaction’(P-0.01,rsy0.35),‘interoceptive awareness’(P-0.01,rsy0.38), and ‘maturity fear’ (P-0.04, rsy0.31). When stepwise multiple regression anal-ysis was applied to assess the influence of eachEDI psychopathological dimension on plasmaestrogen variability, this showed that variability inE values could be accounted for only ‘interoceptive

awareness’(F , Ps0.01,Rs0.18), ‘ineffective-1.33

ness’(F s7.77,Ps0.008,Rs0.16) and ‘inter-1.33

personal distrust’(F s6.19,Ps0.02,Rs0.10).1.33

There were no correlations with B.I.T.E., Y-BOCS-ED or TPQ scores. There were no correlationsbetween LEP and E values.

4. Discussion and conclusions

Our data offer some suggestions about the pos-sible pathogenetic background of persistence ofamenorrhoea after nutritional recovery of ouranorexics.The lack of restoration of menses does not seem

to be due to nutritional deficiency, since BMIvalues at the time of our study and at least in theprevious 3 months were no different in patientsand controls, our anorexics having reached valuesconsidered amply sufficient to resume menstrua-tion according to the data in the literature and tothe critical body weight hypothesis(Frisch andMcArthur, 1974; Frisch, 1977; Mazure et al.,1994). Even though we could not investigate therelative percentages of lean and fat mass in ourpatients, and therefore exclude that they had incor-rect proportions of the two masses possibly respon-sible for the persistence of the amenorrhoea(Wentz, 1982; Mathiak et al., 1999), the stable,well-balanced eating behavior attained by ouranorexics tends to suggest that a correct proportionbetween the two types of tissues should have beenreached. Moreover, a recent study(Kerruish et al.,2002) demonstrated a significant correlationbetween BMI values and percentage of body fat(examined by dual-energy X-ray absorptiometry)in both severely malnourished anorexics andhealthy controls.Among the psychological parameters, there were

still some alterations in our weight-recovered anor-exics. In agreement with the data of Frisch(1977)and Meyer et al.(1986a), we observed higher thannormal B.I.T.E. and E.D.I.-1 scores and, amongthe latter in particular, those related to ‘perfection-ism’, ‘drive for thinness’, ‘body dissatisfaction’and ‘interoceptive awareness’, which are all coresymptoms of AN. Moreover, our patients still hadhigher than normal HA and lower than normal SSscores, as expected, since these are trait parameters


reported to occur in anorexics(Casper, 1990). Thenormal RD scores, instead, were not expected sincethey are also trait parameters and have beenreported to be higher than normal in AN. Thepersistence of some psychopathological aspectstypical of AN after normalization of feeding andweight recovery suggests that they do not resultonly from malnutrition, but may represent thepsychological background for the full developmentof the disease, and possibly for its persistence andrelapse.The data on the hormonal functions are intrigu-

ing. While there seemed to be no pathology forthe hypothalamo–pituitary–adrenal(HPA) and thehypothalamo–pituitary–thyroid axes, the study ofHPG activity revealed that while the baselinelevels of the two gonadotropins were normal, Econcentrations were still definitely lower thanwould be expected in view of the stabilized nutri-tional state of the patients. The dissociationbetween pituitary and ovarian function is intrigu-ing, though it could be a misleading result, sincewe measured only basal secretions of FSH and LHand not their pulsatility or cyclical secretions,which are essential for the stimulation of ovarianfunction(Katz et al., 1978). The lower than normalE secretion could not be due to AN linked persist-ing nutritional alterations, since it did not correlatewith BMI values, with thyroid insufficiency,excluded by the normal values of the stimulatoryTSH, T and T values, or with hyperactivity of4 3

the inhibitory HPA system, excluded by the normal24 h free CORT values.The second intriguing hormonal alteration was

the significantly lower than normal LEP values. Itis well known that LEP is a hormonal product ofadipocytes and a peripheral signal of the nutritionalstatus to the central nervous system. In AN, LEPvalues are extremely low throughout the course ofthe disease, then increase to higher than normalvalues when weight recovery is reached, and returnto normal thereafter(Greenspoon et al., 1996;Hebebrand et al., 1997; Letiexhe et al., 1997;Licinio, 1997; Kopp et al., 1997; von Prittwitz etal., 1997; Baranowska et al., 1997; Mantzoros etal., 1997; Balligand et al., 1998; Eckert et al.,1998; Herpertz et al., 1998, 2000; Audi et al.,1998; Morgan et al., 1999; Gendal et al., 1999;

Ceci et al., 2001; Wabitsch et al., 2001). In ourpatients we expected higher than normal or normalvalues, not the extremely low concentrationsobserved. The lack of correlation between LEPand present or previous nadir BMI values tends toexclude that the phenomenon is related to nutrition,even though the observation that there was anegative correlation between LEP values and thedifference between the BMI levels before thebeginning of AN and those at the moment of ourstudy is intriguing. It could be that even thoughthe present BMIs for our cases were perfectlynormal, they might have been perceived metabol-ically as still not correct since they had not reachedthe pre-AN levels, which were relatively highereven though still normal, and could possibly rep-resent the metabolic optimum for these patients.This hypothesis has been previously suggested byCrisp and Stonehill(1971), even though not defi-nitely confirmed. The only sound correlation withthe LEP deficiency seems to be the negative onewith the E.D.I.-1 psychopathological parameters.This is a totally new observation, since interrelationships between LEP secretion and psycho-logical or behavioral aspects, apart from thoserelated to eating, have never been reported. Wehave no explanation for this phenomenon, sincethe data regarding the biological bases of theanorexic psychopathology are too scanty to enableus to suggest a possible direct role for LEP inpsychological function, or indirect ones on neuro-transmitters, neuropeptides or neurohormones thatregulate both LEP secretion and psychological-behavioral parameters.The fact that E deficiency is also correlated

negatively with the persistence of E.D.I.-1 psycho-pathological parameters, some of them overlappingwith those correlating with LEP levels, is intrigu-ing. In experimental animals and humans, LEPstimulates the HPG axis and therefore E secretion,while it may directly inhibit ovarian E and PROGsecretion(Wauters et al., 2000; Mystowski andSchwartz, 2000). In turn, LEP secretion is stimu-lated in vitro by E, and in vivo in ovariectomizedanimals is reduced and returns to normal by Esupplementation. In humans, FSH stimulates LEPand E secretion in parallel(Shimizu et al., 1997;Mannucci et al., 1998; Kristensens et al., 1999).


Therefore, a relationship between the secretions ofthe two hormones has been described repeatedly.However, which of the two impairments, LEP orE deficiency, comes first in our anorexics is diffi-cult to say, especially because we saw no signifi-cant correlation between the two alterations. Thefact that both correlate with the same psychopath-ological parameters suggests that there must be alink between the two alterations.In conclusion, specific hormonal and psycho-

pathological impairments go in parallel and pos-sibly underlie the amenorrhoea that persists afterweight recovery in AN, and may represent patho-genetic factors for its chronicity or relapse.

References

American Psychiatric Association, 1994. DSM IV. Diagnosticand Statistical Manual of Mental Disorder. fourth ed..American Psychiatric Press, Washington, DC.

Audi, L., Mantzoros, C.S., Vidal-Puig, A., Vargas, D., Gussi-nye, M., Carrascosa, A., 1998. Leptin in relation to resump-tion of menses in women with anorexia nervosa. MolecularPsychiatry 3, 544–547.

Balligand, J.L., Brichard, S.M., Brichard, V., Desager, J.P.,Lambert, M., 1998. Hypoleptinemia in patients with anorex-ia nervosa: loss of circadian rhythm and unresponsivenessto short-term refeeding. European Journal of Endocrinology138, 415–420.

Baranowska, B., Wasileska-Dziubinska, E., Radzikowska, M.,Plonowski, A., Roguski, K., 1997. Neuropeptide Y, galanin,and leptin release in obese women and in women withanorexia nervosa. Metabolism 46, 1384–1389.

Bell, E.T., Harkness, R.A., Loraine, J.A., Russell, G.F.M.,1966. Hormone assay studies in patients with anorexianervosa. Acta Endocrinologica 51, 140–148.

Beumont, P.J.V., George, G.C.W., Pimstone, B.L., Vinik, A.,1976. Body weight and pituitary response to hypothalamicreleasing hormones in patients with anorexia nervosa. Jour-nal of Clinical Endocrinology and Metabolism 43, 487–496.

Buvat, J., Buvat-Herbaut, M., Lemaire, A., Racadot, A.,Fourlinnie, J.C., 1984. Comparison of estrogen primingeffects with body weight restoration effects on the gonado-tropin pattern of patients with anorexia nervosa. HormoneResearch 20, 224–230.

Casper, R.C., 1990. Personality features of women with goodoutcome from restricting anorexia. Psychosomatic Medicine52, 156–170.

Ceci, L., Covezzi, R., Orlandi, E., Guaraldi, G.P., 2001.Correlazione della concentrazione serica di leptina convariabili antropometriche, neuroendocrine, psicopatologichein soggetti con disturbi dell’alimentazione. Eating andWeight Disorders 6, 12.

Cloninger, C.R., Przybeck, T.R., Svrakic, D.M., Wetzel, R.D.,1994. The Temperament and Character Inventory: A Guideto its Development and Use. Center for Psychobiology ofPersonality, St. Louis, MO.

Crisp, A.H., Stonehill, E., 1971. Relation between aspects ofnutritional disturbance and menstrual activity in primaryanorexia nervosa. British Medical Journal 3, 149–151.

Dally, P.J., Sargant, W., 1966. Treatment and outcome ofanorexia nervosa. British Medical Journal ii, 793–795.

Eckert, E.D., Pomeroy, C., Raymond, N., Kohler, P.F., Thuras,P., Bowers, C.Y., 1998. Leptin in anorexia nervosa. Journalof Clinical Endocrinology and Metabolism 83, 791–795.

Falk, R.M., Halmi, K.A., 1982. Amenorrhea in anorexianervosa: examination of the critical body weight hypothesis.Biological Psychiatry 17, 799–806.

Frisch, R.E., McArthur, J.W., 1974. Menstrual cycles: fatnessas a determinant of minimum weight for height necessaryfor their maintenance or onset. Science 185, 949–951.

Frisch, R.E., 1977. Food intake, fatness and reproductiveability. In: Vigersky, R. (Ed.), Anorexia Nervosa. NewYork, Raven Press, pp. 149–161.

Garner, D.M., Olmstead, M.A., Polivy, J., 1983. Developmentand validation of a multidimensional eating disorder inven-tory for anorexia nervosa and bulimia. International Journalof Eating Disorders 2, 15–34.

Gendal, K.A., Kaye, W.H., Altemus, M., McConaha, C.,La Via, M.C., 1999. Leptin, neuropeptide Y, and peptideYY in long-term recovered eating disorders patients. Bio-logical Psychiatry 46, 292–299.

Greenspoon, S., Gulick, T., Askari, H., Landt, M., Lee, K.,Anderson, E., Ma, Z., Vignati, L., Bowsher, R., Herzog, D.,Klibanski, A., 1996. Serum leptin levels in women withanorexia nervosa. Journal of Clinical Endocrinology andMetabolism 81, 3861–3863.

Hebebrand, J., Blum, W.F., Barth, N., Coners, H.,Englarop Juul, A., Ziegler, A., Warnke, A., Rascher, W.,Remschmidt, H., 1997. Leptin levels in patients with ano-rexia nervosa are reduced in the acute stage and elevatedupon short-term weight restoration. Molecular Psychiatry 2,330–334.

Henderson, M., Freeman, C.P.L., 1987. Bulimic InvestigationTest Edinburgh(B.I.T.E.). A self-rating scale for bulimia:the BITE. British Journal of Psychiatry 150, 18–24.

Herpertz, S., Wagner, R., Albers, N., Blum, W.F., Pelz, B.,Langkafel, M., Kopp, W., Henning, A., Oberste-Berghaus,C., Mann, K., Senf, W., Hebebrand, J., 1998. Circadianplasma leptin levels in patients with anorexia nervosa:relation to insulin and cortisol. Hormone Research 50,197–204.

Herpertz, S., Albers, N., Wagner, R., Pelz, B., Kopp, W.,Mann, K., Blum, W.F., Senf, W., Hebebrand, J., 2000.Longitudinal changes of circadian leptin, insulin and cortisolplasma levels and their correlation during refeeding inpatients with anorexia nervosa. European Journal of Endo-crinology 142, 373–379.


Jamrozik, S., Dumoulin, S., Saint-Martin, F., de Glisezinski,I., Barbe, P., Plantavid, M., Thouvenot, J.P., Bennet, A.,Louvet, J.P. 1995. LH secretion in anorectic women whoremain amenorrheic after partial or total recovery of bodyweight. 15th Congress of the International Federation Fer-tility Society, Montpellier, France(abstract).

Katz, J.L., Boyar, R., Roffwarg, H., Hellman, L., Weiner, H.,1978. Weight and circadian luteinizing hormone secretorypattern in anorexia nervosa. Psychosomatic Medicine 40,549–567.

Kerruish, K.P., O’Connor, J., Humphries, J.R.J., Kohn, M.R.,Clarke, S.D., Briody, J.N., Thomson, E.J., Wright, K.A.,Gaskin, K.J., Baur, L.A., 2002. Body composition in ado-lescents with anorexia nervosa. American Journal of ClinicalNutrition 75, 31–37.

Kohmura, H., Miyake, A., Aono, T., Tanizawa, O., 1986.Recovery of reproductive functions in patients with anorexianervosa: a 10-year follow-up study. European Journal ofObstetrics, Gynecology and Reproductive Biology 22,293–296.

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 levelspredict amenorrhea in underweight and eating disorderedfemales. Molecular Psychiatry 2, 335–340.

Kristensens, K., Pedersen, S.B., Richelsen, B., 1999. Regula-tion of leptin by steroid hormones in rat adipose tissue.Biochemical and Biophysical Research Communication 259,624–630.

Lecomte, P., Kabir-Gros, N., Lansac, J., 1984. La fonctiongonadique des anorexies nerveuse a la phase post critique.Ànnales d’Endocrinologie(Paris) 45, 397–401.

Letiexhe, M.R., Scheen, A.J., Lefebvre, P.J., 1997. Plasma`leptin levels, insulin secretion, clearance and action onglucose metabolism in anorexia nervosa. International Jour-nal of Eating and Weight Disorders 2, 79–86.

Licinio, J., 1997. Leptin in anorexia nervosa and amenorrhea.Molecular Psychiatry 2, 267–269.

Mannucci, E., Ognibene, A., Becorpi, A., Cremasco, F., Pel-legrini, S., Ottanelli, S., Rizzello, S.M., Massi, G., Messeri,G., Rotella, C.M., 1998. Relationship between leptin andoestrogens in healthy women. European Journal of Endocri-nology 139, 198–201.

Mantzoros, C., Flier, J.S., Lesem, M.D., Brewerton, T.D.,Jimerson, D.C., 1997. Cerebrospinal fluid leptin in anorexianervosa: correlation with nutritional status and potential rolein resistance to weight gain. Journal of Clinical Endocrinol-ogy and Metabolism 82, 1845–1851.

Mathiak, K., Gowin, W., Hebebrand, J., Ziegler, A., Blum,F.W., Felsenberg, D., Lubbert, H., Kopp, W., 1999. Serumleptin levels, body fat disposition, and weight in femaleswith anorexia or bulimia nervosa. Hormone and MetabolicResearch 31, 274–277.

Mazure, C.M., Halmi, K.A., Sunday, S.R., Romano, S.J.,Einborn, A.M., 1994. The Yale-Brown-Cornell Eating Dis-order Scale: development, use, reliability and validity. Jour-nal of Psychiatric Research 28, 425–445.

Meyer, A.E., von Holtzapfel, B., Deffner, G., Engel, K., Klick,M., 1986a. Amenorrhea and predictors of remenorrhea inanorexia nervosa: a psychoendocrinological study in inpa-tients. Psychotherapy and Psychosomatics 45, 149–160.

Meyer, A.E., von Holtzapfel, B., Deffner, G., Engel, K., Klick,M., 1986b. Psychoendocrinology of remenorrhea in the lateoutcome of anorexia nervosa. Psychotherapy and Psycho-somatics 45, 174–185.

Morgan, J.F., Bollon, J., Sedwick, P.M., Patel, S., Lacey, J.H.,Conway, G.S., 1999. Changes in plasma concentrations ofleptin and body fat composition during weight restorationin anorexia nervosa. Journal of Clinical Endocrinology andMetabolism 84, 1999–2107.

Mystowski, P., Schwartz, M.W., 2000. Gonadal steroids andenergy homeostasis in the leptin era. Nutrition 16, 937–946.

Nakamura, Y., Yoshimura, Y., Oda, T., Katayama, E., Kamey,K., Tanabe, K., Iizuka, R., 1985. Clinical and endocrinestudies on patients with amenorrhoea associated with weightloss. Clinical Endocrinology 23, 643–651.

Russell, G.F.M., 1972. Psychological and nutritional factors indisturbance of menstrual function and ovulation. Postgrad-uate Medical Journal 48, 10–13.

Shimizu, H., Shimomura, Y., Nakanishi, Y., Futawatari, T.,Ohtani, K., Sato, N., Mori, M., 1997. Estrogen increases invivo leptin production in rats and human subjects. Journalof Endocrinology 154, 285–292.

Spitzer, R.L., 1994. Structured Clinical Interview for DSM-IV. American Psychiatric Association, Washington, DC.

Treasure, J.L., Wheeler, M., King, E.A., Gordon, P.A.L.,Russell, G.F.M., 1988. Weight gain and reproductive func-tion: ultrasonographic and endocrine features in anorexianervosa. Clinical Endocrinology 29, 607–616.

Valcke, J.C., Mahoudeau, J.A., Thieblot, P.H., Pique, L., Luton,J.P., Franchilont, P., Moreau, L., Bricaire, H., 1974. Criteres`d’interpretation du test utilisant l’hormone de liberation de´ ´la luteostimuline(LHRH). Annales d’Endocrinologie(Par-ís) 35, 423–443.

von Prittwitz, S., Blum, W.F., Ziegler, A., Scharmann, S.,Remschmidt, H., Hebebrand, J., 1997. Restrained eating isassociated with low leptin levels in underweight females.Molecular Psychiatry 2, 420–422.

Wabitsch, M., Ballauff, A., Holl, R., Blum, W.F., Heinze, E.,Remschmidt, H., Hebebrand, J., 2001. Serum leptin, gonad-otropin, and testosterone concentrations in male patientswith anorexia nervosa during weight gain. Journal of Clin-ical Endocrinology and Metabolism 86, 2982–2984.

Wackeling, A., Marshall, J.C., Beardwood, C.J., De Souza,V.A., Russell, G.F.M., 1976. The effects of clomiphenecitrate on the hypothalamic-pituitary-gonadal axis in anorex-ia nervosa. Psychological Medicine 6, 371–380.

Wackeling, A., De Souza, V.A., Beardwood, C.J., 1977.Assessment of the negative and positive effects of admin-istered oestrogen on gonadotropin release in patients withanorexia nervosa. Psychological Medicine 7, 397–405.


Warren, M.P., Jewelewicz, R., Dyrenfurth, I., Ans, R., Khalaf,S., Vande Wiele, R.L., 1975. The significance of weightloss in the evaluation of pituitary response to LHRH inwomen with secondary amenorrhoea. Journal of ClinicalEndocrinology and Metabolism 40, 601–611.

Wauters, M., Considine, R.W., Van Gaal, L.F., 2000. Humanleptin: from an adipocyte hormone to an endocrine mediator.European Journal of Endocrinology 143, 293–311.

Wentz, A.C., 1982. Body weight and amenorrhoea. Obstetricsand Gynecology 56, 482–487.

Documents

Persistent amenorrhoea in weight-recovered anorexics: psychological and biological aspects