P

Tap

CY

a

Kb

Kc

d

e

f

R

T

h0

sychoneuroendocrinology (2015) 56, 52—61

Available online at www.sciencedirect.com

ScienceDirect

jou rn al h om epage: www.elsev ier .com/ locate /psyneuen

he late-luteal leptin level, caloric intakend eating behaviors among women withremenstrual dysphoric disorder

hih-Hung Koa,b,c, Cheng-Fang Yenb,c, Cheng-Yu Longd,u-Ting Kuoe, Cheng-Sheng Chenb,c, Ju-Yu Yenb,c,f,∗

Department of Psychiatry, Kaohsiung Municipal Hsiao-Kang Hospital, Kaohsiung Medical University,aohsiung, TaiwanDepartment of Psychiatry, Faculty of Medicine, College of Medicine, Kaohsiung Medical University,aohsiung City, TaiwanDepartment of Psychiatry, Kaohsiung Medical University Hospital, Kaohsiung, TaiwanDepartment of Obstetrics and Gynecology, Kaohsiung Medical University Hospital, Kaohsiung, TaiwanDepartment of Radiology, Chi-Mei Medical Center, Tainan, TaiwanDepartment of Psychiatry, Kaohsiung Municipal Ta-Tung Hospital, Kaohsiung, Taiwan

eceived 19 October 2014; received in revised form 28 February 2015; accepted 1 March 2015

KEYWORDSPremenstrualdysphoric disorder(PMDD);Caloric intake;Leptin;Uncontrolled eating

SummaryObjectives: A marked increased in food intake in the late-luteal phase is a characteristic symp-tom of premenstrual dysphoric disorder (PMDD). The aim of the study was to evaluate theleptin level, caloric intake, and eating behaviors of women with PMDD across the menstrualcycle among normal-weight and overweight subjects, respectively.Methods: A total of 62 women with PMDD and 69 controls were recruited following psychiatricinterviewing and underwent prospective investigation. The leptin level, caloric intake, andthree factors related to eating behavior were assessed in both the late-luteal and follicularphases.Results: The women with PMDD had greater increases in caloric intake, sweet caloric intake,and uncontrolled eating in the late-luteal phase than the controls. Among the normal-weight

women, the leptin level was negatively correlated with caloric intake. The normal-weightwomen with PMDD had a lower leptin level, a lower leptin/body fat percentage (BFP), a highercaloric intake, and higher uncontrolled eating and emotional eating in the late-luteal phasethan the normal-weight controls. Their leptin level was correlated negatively with sweet caloricintake. On the other hand, the overweight women with PMDD had a higher leptin level thanthe normal-weight women with PMDD and no decline was observed in their leptin level in the

∗ Corresponding author at: Department of Psychiatry, Kaohsiung Medical University Hospital, 100, Tzyou 1st Road, Kaohsiung 80708,aiwan. Tel.: +886 7 3121101x6822.

E-mail address: yenjuyu@cc.kmu.edu.tw (J.-Y. Yen).

ttp://dx.doi.org/10.1016/j.psyneuen.2015.03.002306-4530/© 2015 Elsevier Ltd. All rights reserved.

Leptin in PMDD

late-luteal phase. There were no

with PMDD and the overweight councontrolled eating and emotionaConclusions: A decline in the leptweight women with PMDD. Hyperlwith PMDD should be monitored anThe detailed roles of leptin and oovereating among women with PM© 2015 Elsevier Ltd. All rights res

racelm2n(dwitnt(

bareuth2cabotspc

1f

AcCt2ms2014). Further, women with PMDD have a higher impulsivity

1. Introduction

Premenstrual dysphoric disorder (PMDD) was included as anofficial diagnostic criteria in the DSM 5 in 2012 (AmericanPsychiatric Association, 2012) based on adequate evidence(Epperson et al., 2012). It is defined as having predictable,cyclic, and functional-impairing psychological and somaticsymptoms, such as depression, irritability, lethargy or hyper-somnia. These symptoms are aggravated in the late-lutealphase of the menstrual cycle and resolved by menstrua-tion (Rapkin and Winer, 2009). Marked changes in appetiteand overeating are typical concomitant PMDD symptoms(American Psychiatric Association, 2012) and are affectedby the menstrual cycle among women with PMDD (Yen et al.,2010). Evaluation of hormones that fluctuate during themenstrual cycle and affect eating behavior, such as lep-tin, may provide insight into the mechanism of overeatingsymptoms among women with PMDD.

1.1. The caloric intake of women with PMDD

Food intake is reduced in the periovulatory phaseand increased during the luteal phase among women(Buffenstein et al., 1995; Dye and Blundell, 1997; Van Vugt,2010), particularly among those with PMDD (Evans et al.,1999; McNeil and Doucet, 2012). Women with PMDD havebeen reported to eat more calories in the late-luteal phasethan in the follicular phase as compared with control women(Reed et al., 2008). Further, sweet food craving is exacer-bated in the late-luteal phase among women with PMDD (Yenet al., 2010). These results indicated that women with PMDDincrease their food intake in the late-luteal phase. How-ever, whether women with PMDD have an increased caloricintake and an increased sweet caloric intake in the late-luteal phase as compared with healthy women has not beenempirically evaluated throughout the menstrual cycle. Addi-tionally, the underlying factors contributing to the changesin caloric intake among women with PMDD are not well-understood (McNeil and Doucet, 2012).

1.2. The role of leptin in the appetite of womenwith PMDD

Leptin, a hormone secreted by adipocytes, acts on thehypothalamus to regulate appetite and neuroendocrine

function (Cohen, 2006). Leptin has a role in ovulationthrough its stimulation of the pituitary to secrete luteinizinghormone (LH) (Fenichel et al., 2008). The administration of

(eu

53

differences in the leptin level between the overweight womenntrols. They also had a higher sweet caloric intake and higherl eating than the overweight controls.in level is associated with late-luteal overeating among normal-eptinemia and a high sweet caloric intake of overweight womend addressed in order to attenuate the risk of leptin resistance.ther psycho-neuro-endocrinology factors in the mechanism ofDD should be evaluated in future study.

erved.

ecombinant leptin has been shown to restore mensesnd fertility in women with hypothalamic or leptin defi-iency amenorrhea (Chou et al., 2011; von Schnurbeint al., 2012). Recent studies have demonstrated that theeptin level is increased over the menstrual cycle, with aid-cycle peak concurrent with the LH surge (Ahrens et al.,

014). These studies all indicated that leptin is related toot only eating behavior, but also to the menstrual cycleGoumenou et al., 2003). One study revealed a significantecline in leptin among women with PMDD in comparisonith controls (Akturk et al., 2013). As leptin plays key roles

n the menstrual cycle and appetite and has been foundo be altered among women with PMDD, further study isecessary to evaluate the alterations in the leptin level andheir effects on the caloric intake of women with PMDDAkturk et al., 2013).

Further, leptin is transported across the blood-brain-arrier by a saturable transporter system and exerts itsnorectic effect via the hypothalamic arcuate nucleus. Itesults in a reduced food intake and an increased energyxpenditure (Suzuki et al., 2012). However, obese individ-als often have high leptin levels, which result in a failureo respond to exogenous leptin (Suzuki et al., 2012). Obesityas been reported to be associated with PMDD (Masho et al.,005). The moderating effect of being overweight should beonsidered when evaluating the association between leptinnd caloric intake. Further, there exist complex associationsetween the menstrual cycle, leptin level, eating behavior,besity, and PMDD (McNeil and Doucet, 2012), and evalua-ion of these factors across the menstrual cycle with analysiseparately in overweight and normal-weight women mayrovide insight into the effect of leptin on the late-lutealaloric intake among women with PMDD.

.3. The role of PMDD symptoms and associatedactors in caloric intake in the late-luteal phase

side from hormone levels, mood, stress, and rewardingharacteristics contribute to overeating behavior (Jauch-hara and Oltmanns, 2014). Depression and irritability arehe cardinal symptoms of PMDD (Ko et al., 2013; Yen et al.,011). Emotional eating is an eating behavior motivated byultiple negative emotions (Chesler, 2012) and moderate

tress-induced food intake in females (van Strien et al.,

Yen et al., 2011). Impulsivity is associated with uncontrolledating (Leitch et al., 2013), a tendency to eat more thansual due to a loss of control over intake (Karlsson et al.,

5

22taomhwtrtw

1

Ttpo(bebaaw

2

2

SsatreAtipt1tlSowpysawtdvDEtrt

eR

2

Tw2aTsPft

Qes1coerbw

dotdfmTiotpdai

sas(2T

fshdo((

w

4

000), and contributes to caloric intake (Kronick et al.,011). Women with PMDD also have impaired inhibitory con-rol (Yen et al., 2013). The inhibitory control function isssociated with cognitive restraint, conscious restrictionf food intake in order to control body weight or to pro-ote weight loss (Leitch et al., 2013). Cognitive restraint

as been reported to associate with caloric intake amongomen with PMDD (Evans et al., 1999). Thus, whether

hree factors related to eating behavior, including cognitiveestraint, uncontrolled eating and emotional eating, con-ribute to caloric intake should be evaluated among womenith PMDD.

.4. The aims of the study

hus, the aims of the study were to evaluate: (1) whetherhe menstrual cycle (late-luteal phase versus follicularhase) or PMDD (PMDD group versus controls) have impactsn the leptin level, caloric intake, and eating behaviors;2) whether being overweight moderates the associationetween leptin level and caloric intake; (3) the differ-nces in leptin level, caloric intake and eating behaviorsetween women with PMDD and controls among overweightnd normal-weight subjects; (4) whether eating behaviorsre associated with late-luteal caloric intake among normal-eight or overweight women with PMDD.

. Methods

.1. Participants

ubjects were recruited using an advertisement postedpecifically to attract a PMDD group without treatmentnd a control group on a university campus from 08/2011o 10/2012. Volunteers in the PMDD group had a positiveesponse to five or more premenstrual symptoms of theleven DSM-IV-TR criteria of PMDD (American Psychiatricssociation, 2000). The control group had positive responseso fewer than 2 of those criteria or had no functionalmpairment under mild symptoms. Subjects currently takingsychotropic or hormone medication, such as contracep-ive medication, were excluded from the study. A total of69 (90 PMDD candidate subjects and 79 candidate con-rol subjects) women with a college or higher educationalevel were screened using the self-reported Premenstrualymptoms Screening Tool (PSST) after informed consent wasbtained. Eighty-six participants of the PMDD candidatesere screened and were found to have moderate to severeremenstrual symptoms and entered the next steps of anal-sis. In addition, 75 women of the candidate controls werecreened and found not to have premenstrual symptomsnd entered the next steps of analysis. The other subjectsere excluded. The women were then interviewed by one of

wo psychiatrists to exclude psychotic disorder and bipolar Iisorder using the Mini-International Neuropsychiatric Inter-iew (MINI), and to make the diagnosis of PMDD based on theSMIV-TR criteria (American Psychiatric Association, 2000).

ighty-three PMDD candidate subjects and 75 candidate con-rols were diagnosed as having PMDD and not having PMDD,espectively. All of the subjects participated in a prospec-ive investigation to confirm their premenstrual symptoms

lfta

C.-H. Ko et al.

xacerbation. The study was approved by the Institutionaleview Board (IRB) of Kaohsiung Medical University Hospital.

.2. Measures

he Premenstrual Symptoms Screening Tool (PSST). The toolas developed by Steiner and colleagues (Steiner et al.,003). We utilized the scale to screen women with moder-te to severe premenstrual symptoms (Steiner et al., 2003).hey had to have (1) at least 1 of the first 4 items of the 14ymptoms; (2) four or more items of the 14 items assessingMDD symptoms; and (3) at least 1 of the 5 items assessingunctional impairments that were classified as of a moderateo severe level, with a score of 3 or 4.

The Premenstrual Dysphoric Disorder (PMDD) Severityuestionnaire (PMDDSQ). The PMDDSQ was developed tovaluate the severity of PMDD symptoms across the men-trual cycle (Yen et al., 2013). The questionnaire is an1-point Likert questionnaire rating the severity of the 11riteria of PMDD in the DSM IV-TR. Every symptom was ratedn a scale ranging from 0 to 10 to represent none at all to anxtreme severity. The total score of the 11 items was used toepresent the PMDD severity. The questionnaire had a Cron-ach’s alpha of 0.98 and the four-week test-retest reliabilityas 0.92.

The mean caloric intake. The subjects recorded theirietary intake for three days (the two last weekdays andne last holiday in the week before menstruation and inhe week after menstruation). They recorded all food andrink consumed, except water. The eating time, name of theood (including trade mark), estimated volume or weight,ethod of cooking and sugar levels of drinks were noted.he caloric values of the foods were calculated according to

nformation from the Food and Drug Administration, Ministryf Health and Welfare, Taiwan. The total caloric intake washen divided by three to determine the mean caloric intakeer day. Sweet-tasting foods, such as cake, ice cream, candy,oughnuts, waffles, cookies or chocolate, and sugary bever-ges were classified as sweet food. The sweet food caloricntake was calculated in the same way.

18-item Three-Factor Eating Questionnaire (TFEQ). Theelf-assessed four-point response scale measures cognitivend behavioral components of eating. It contains three sub-cales, cognitive restraint (6 items), uncontrolled eating9 items), and emotional eating (3 items) (Karlsson et al.,000). The Cronbach’s ˛ values of the 3 subscales of theFEQ were 0.77, 0.83, and 0.86, respectively.

Body mass index (BMI) is a measure of weight adjustedor height, calculated as weight in kilograms divided by thequare of height in meters (kg/m2). Subjects with a BMIigher than 24 were classified as overweight based on theefinition of the Health Promotion Administration, Ministryf Health and Welfare, Taiwan. Further, body fat percentageBFP in %) was assessed using a body composition monitor@Omron).

The levels of estrogen and progesterone. Blood samplesere obtained directly from all subjects from a cannu-

ated vein after testing in the late-luteal phase and theollicular phase. The serum was separated by centrifuga-ion and stored at −20 ◦C until further analysis. Estrogennd progesterone were measured using a Coat-A-count

3

Awlstws1bsPatdgcew

3i

RcFuiFtTepeb

3cn

ATlhltw(iscwrb

Leptin in PMDD

radioimmunoassay kit (Siemens Medical SolutionsDiagnostics, Los Angeles, CA, USA).

The level of leptin. Serum and plasma samples werestored at −20 ◦C until analysis. Serum leptin levels weremeasured using an RIA kit from Linco Research, a divisionof Millipore Inc. (Cat# HL-81K, St Charles, MO, USA). Theintra-assay CV ranged from 3.4 to 8.35% and the inter-assayCV ranged from 3.0 to 6.2%. The sensitivity was 0.5 ng/mL.The assays were run on a Cobra II series auto-gamma count-ing system (Packard Instrument Company, Meriden, CT, USA).The serum leptin level was standardized for BFP (dividing theserum leptin level by BFP; leptin/BFP) to control the effectof body fat percentage.

2.3. Procedures

All participants were evaluated in both the late-luteal phase(within one week before menstruation as predicted by thelast menstruation cycle) and the follicular phase (within oneweek after the end of menstruation and before ovulating).All subjects were prohibited from eating or drinking, withthe exception of water in a volume lower than 300 cm3, inthe three hours before entering the laboratory. After 1.5 h ofquestionnaire completion and cognitive assessment, bloodwas obtained to assess the estrogen, progesterone, and lep-tin levels. The participants were allowed to eat after theblood test. Half of the subjects (39 in the PMDD group and39 in the control group) were assessed first in the late-lutealphase and then again in the follicular phase. The other sub-jects were assessed in reverse. After the investigation in thefirst menstrual cycle, all participants were followed-up andcompleted the PMDDSQ once a week for 2 menstrual cycles.

2.4. Data analysis

The differences in age, educational level, BMI, BFP, estro-gen, and progesterone between the PMDD and control groupwere evaluated by the t test. Repeated-measures two-factor analysis of variance (ANOVA) was used to evaluatethe leptin level, caloric intake, and eating behaviors as afunction of the menstrual cycle phase (late-luteal phaseversus follicular phase) and PMDD diagnosis (PMDD groupversus control group), controlling for age and educationallevel, to demonstrate the late-luteal changes in PMDD. Asleptin and leptin/BFP were not of a normal distribution,non-parametric analysis was used to evaluate the leptinlevel data. We also evaluated the difference in leptin levelbetween the overweight group and the normal-weight groupusing the Mann—Whitney Test. Two-factor ANOVA was usedto evaluate the leptin level as a function of PMDD diagnosis(PMDD group versus control group) and weight (overweightgroup versus normal-weight group). Then, the associationbetween the leptin level and PMDD was evaluated separatelyin the overweight and normal-weight groups. A p-valuelower than 0.05 was considered statistically significant forthe above analyses. The correlations between leptin level

and caloric intake were evaluated by Spearman’s analysis. Ap-value lower than 0.0125 was considered statistically sig-nificant for Spearman’s analysis after Bonferroni correctionin multiple tests.

lets

55

. Results

ccording to the definition of a symptomatic cycle in PMDD,omen with PMDD should have a PMDDSQ score in the late-

uteal phase that is 30% higher than the individual minimalcore in the menstrual cycle (Smith et al., 2003). Further,he score on the PMDDSQ in the late-luteal phase amongomen with PMDD should be higher than the mean plus two

tandard deviations of that in the control group. A total of6 PMDD candidates were excluded from the PMDD groupecause they did not meet the criteria for two consecutiveymptomatic cycles (Smith et al., 2003). Five women withMDD and 6 controls did not complete the caloric intakessessment. A total of 62 women with PMDD and 69 con-rols were entered into the final analysis. There were noifferences in age, educational level, estrogen level or pro-esterone level between the women with PMDD and theontrols (Table 1). There were also no significant differ-nces in BMI or body fat percentage between the womenith PMDD and the controls.

.1. The late-luteal changes in leptin level, caloricntake and eating behaviors (Fig. 1)

epeated-measures two-factor ANOVA demonstrated signifi-ant late-luteal increases in caloric intake (F = 6.21, p = 0.01;ig. 1A), sweet caloric intake (F = 4.96, p = 0.03; Fig. 1B), andncontrolled eating (F = 27.78, p < 0.001; Fig. 1F), and signif-cant late-luteal declines in leptin level (F = 4.09, p = 0.045;ig. 1C) and leptin/BFP (F = 7.14, p = 0.008; Fig. 1D) amonghe women with PMDD in comparison with the control group.he interaction effect was not significant for emotionalating (F = 3.46, p = 0.065) or cognitive restraint (F = 3.34,

= 0.07). There were no significant correlations betweenstrogen or progesterone levels and caloric intake or eatingehaviors.

.2. The correlation between leptin level andaloric intake among the overweight andormal-weight subjects

total of 25 women (19.1%) were classified as overweight.he women who were overweight had a significantly higher

eptin level (late-luteal phase: Z = 5.98, p < 0.001) and aigher leptin/BFP (late-luteal phase: Z = 5.10, p < 0.001) inate-luteal phase. Spearman’s analysis demonstrated thathe late-luteal caloric intake was negatively correlatedith the leptin level (� = −0.32, p = 0.001) and leptin/BFP

� = −0.34, p < 0.001) in the normal-weight group, but notn the overweight group (Table 2). Uni-variable analy-is of covariance was utilized to evaluate the late-lutealaloric intake as a function of leptin level and being over-eight (overweight group versus normal-weight group). This

evealed a significant interaction between the effects ofeing overweight and the leptin level effect on the late-

uteal caloric intake (F = 4.60, p = 0.034). The moderatingffect of being overweight on the association between lep-in level and caloric intake suggested that this associationhould be evaluated separately by body weight.

56 C.-H. Ko et al.

Table 1 Comparisons of demographic data, body mass index (BMI), body fat percentage (BFP), estrogen, and progesteronebetween the women with premenstrual dysphoric disorder (PMDD) and the controls.

Variables (missing N) Mean ± SD Independent t-test

PMDD group (N = 62) Control group (N = 69)

Age 23.35 ± 3.07 23.58 ± 3.51 −0.389Educational level 16.13 ± 1.15 16.25 ± 1.76 −0.446BMI

Late-luteal 21.75 ± 3.48 21.87 ± 3.59 −0.193Follicular 21.86 ± 3.83 21.75 ± 3.49 0.171

Body fat percentage (BFP)Late-luteal 26.80 ± 5.33 27.63 ± 4.45 −0.969Follicular 26.27 ± 4.94 27.89 ± 4.49 −1.970

EstrogenLate-luteal 82.74 ± 6.27 77.05 ± 6.40 −1.005a

Follicular 51.66 ± 3.60 67.86 ± 6.57 −1.098a

ProgesteroneLate-luteal 9.88 ± 8.99 9.07 ± 8.53 −0.550a

Follicular 0.82 ± 1.23 1.13 ± 1.95 −1.725a

3t

Tei

raP

FilU

a Non-parametric analysis: Z values of the Mann—Whitney Test.

.3. The moderating effect of being overweight onhe association between PMDD and leptin

wo-way ANOVA demonstrated significant interactionffects of being overweight and PMDD on the leptin leveln the late-luteal phase (F = 4.99, p = 0.027; Fig. 1E). This

hPt

igure 1 The interaction between the premenstrual dysphoric disntake, leptin level and uncontrolled eating and the interaction betweeptin level. F: the F value for the interaction effect in two-factor

ncontrolled eating: uncontrolled eating subscale of the 18-item Th

esult suggested that being overweight moderates thessociation between PMDD diagnosis and leptin level.airwise analysis demonstrated that overweight women had

igher leptin level than normal-weight ones among bothMDD group and controls. The interaction also revealed thathe difference in leptin level between the overweight and

order (PMDD) effect and the menstrual cycle effect on caloricen the PMDD effect and the overweight effect on the late-lutealanalysis of variance; Leptin/BFP: leptin/body fat percentage;ree-Factor Eating Questionnaire; *: p < 0.05; ***: p < 0.001.

Leptin in PMDD 57

Table 2 The correlations between serum leptin level and caloric intake in late-luteal phase among women with PMDD, controlsand all subjects.

All subjects PMDD group Control group

CI SCI CI SCI CI SCI

Normal-weight subjectsLeptina −0.32** −0.08 −0.16 −0.32 −0.41** 0.22Leptin/BFPa −0.34*** −0.12 −0.14 −0.36* −0.45** 0.17

Overweight subjectsb

Leptina 0.06 −0.05 −0.04 −0.11 0.19 −0.37Leptin/BFPa −0.06 −0.01 0.16 0.04 −0.10 −0.30

CI: caloric intake; SCI: sweet caloric intake.a As leptin level was not of a normal distribution, all analyses of leptin level were evaluated by Spearman’s correlation (�).b As the number of overweight women with PMDD was low (N = 12), the analyses of overweight subjects were evaluated with Spearman’s

correlation (�).* p < 0.0125.

**

2cAtcn

4

Tl2cTbfeiwocsostsbtcslsw

4

p < 0.01.*** p < 0.001.

normal-weight participants was more significant among thewomen with PMDD than among the controls. Thus, the asso-ciation between PMDD and leptin level should be evaluatedseparately in overweight and normal-weight women.

In the normal-weight group, the women with PMDD had alower leptin level and leptin/BFP and a higher caloric intakethan the controls in the late-luteal phase, but not in thefollicular phase (Table 3). Further, they had higher uncon-trolled eating and emotional eating than the controls in boththe late-luteal and follicular phases. Repeated-measurestwo-factor ANOVA demonstrated an interaction betweenthe menstrual cycle effect and PMDD effect on leptin/BFP(F = 4.53, p = 0.036). Spearman’s analysis demonstrated thatthe leptin level and leptin/BFP were negatively correlatedwith the late-luteal caloric intake in the control group, andleptin/BFP was negatively correlated with the late-lutealsweet caloric intake in the PMDD group (Table 2).

As the number of subjects in the overweight group waslimited, their data were evaluated using non-parametricanalysis. There were no differences in the leptin level andleptin/BFP between the PMDD group and the controls amongover-weight participants. The overweight PMDD group had ahigher late-luteal sweet caloric intake and greater uncon-trolled eating than the overweight controls. They also hadhigher emotional eating in both the late-luteal and follicularphases.

4. Discussion

4.1. The eating behaviors of the women withPMDD

The severity of PMDD symptoms was found to be asso-ciated with increased late-luteal caloric intake (Gianniniet al., 1985). In line with a previous report (Reed et al.,

2008), the women with PMDD exhibited a greater increasein caloric intake in the late-luteal phase than the con-trols. This might contribute to a greater body weight (Yenet al., 2010) or a higher cholesterol level (Hsiao et al.,

w

Hp

011). Further, uncontrolled eating was exacerbated signifi-antly in the late-luteal phase among the women with PMDD.lthough the caloric intake and uncontrolled eating yieldo the menstrual cycle effect (late-luteal phase), no asso-iations between them and estrogen or progesterone wereoted in this study.

.2. The effect of leptin level on eating behavior

he leptin level decreases during starvation, and high leptinevels are associated with a decreased appetite (Klok et al.,007). A higher leptin level was correlated with a loweraloric intake in the normal-weight controls in this study.his correlation was highly significant when controlling forody fat, and supported the effect of leptin in regulatingood intake (Klok et al., 2007; Schulz et al., 2012). How-ver, both a high lipid percentage (Minocci et al., 2000) andnsensitivity to leptin (Myers et al., 2012) are associatedith hyper-leptinemia in obese patients. Thus, the effectf the leptin level on eating behavior could be paradoxi-al and could differ between normal-weight and overweightubjects. Further, aside from the interaction between beingverweight and the leptin effect, the menstrual cycle effecthould be evaluated in order to understand its contribu-ion to eating behaviors (McNeil and Doucet, 2012). In thistudy, being overweight moderated the negative correlationetween leptin level and late-luteal caloric intake. Further,his correlation was significant among the normal-weightontrols, but not among the overweight controls. This resultupported the above claim and suggested that the effect ofeptin on regulating caloric intake in the late-luteal phasehould be evaluated separately in normal-weight and over-eight women.

.3. The late-luteal leptin decline among the

omen with PMDD

ealthy women have a higher level of leptin in the lutealhase (Feng et al., 2013). This could explain why women

58 C.-H. Ko et al.

Table 3 Comparisons of leptin level, caloric intake, and eating behaviors between the women with premenstraul dysphoricdisorder (PMDD) and the control group seperately among the normal-weight and overweight subjects.

In the normal-weight group Mean ± SD

PMDD group (N = 50) Control group (N = 56)

Leptin (ng/ml)Late-luteal 9.47 ± 4.81 12.91 ± 6.93 −2.48*,a

Follicular 10.38 ± 6.29 11.87 ± 7.11 −1.21a

Leptin/BFPLate-luteal 0.37 ± 1.60 0.48 ± 2.32 −2.44*,a

Follicular 0.41 ± 0.23 0.43 ± 0.22 −0.53a

Caloric intake (kcal/day)Late-luteal 1327.63 ± 382.62 1161.17 ± 322.19 2.43*

Follicular 1129.26 ± 284.39 1190.09 ± 328.44 −1.01Sweet caloric intake (kcal/day)

Late-luteal 432.53 ± 264.34 368.13 ± 192.56 1.45Follicular 316.35 ± 172.80 362.53 ± 190.31 −1.30

Cognitive restraintLate-luteal 15.56 ± 3.33 16.02 ± 2.99 −0.75Follicular 19.48 ± 4.27 19.29 ± 3.58 0.25

Uncontrolled eatingLate-luteal 22.76 ± 3.96 17.93 ± 4.02 6.22***

Follicular 20.18 ± 3.78 18.27 ± 3.72 2.6*

Emotional eatingLate-luteal 8.00 ± 2.05 5.96 ± 1.53 5.84***

Follicular 7.42 ± 2.40 5.93 ± 1.53 3.77***

In the overweight group Mean ± SD

PMDD group (N = 12) Control group (N = 13)

Leptin (ng/ml)Late-luteal 25.93 ± 9.68 22.81 ± 7.83 −1.14a

Follicular 25.55 ± 8.25 20.07 ± 8.56 0.10a

Leptin/BFPLate-luteal 0.75 ± 0.26 0.69 ± 0.19 0.35a

Follicular 0.77 ± 0.25 0.59 ± 0.19 0.07a

Caloric intake (kcal/day)Late-luteal 1272.44 ± 237.08 1275.15 ± 335.56 0.22a

Follicular 1154.14 ± 154.22 1076.77 ± 252.80 0.98a

Sweet caloric intake (kcal/day)Late-luteal 414.89 ± 231.08 245.69 ± 153.56 2.10*,a

Follicular 422.28 ± 323.77 282.46 ± 179.25 1.03a

Cognitive restraintLate-luteal 15.75 ± 3.17 16.77 ± 2.45 −1.12a

Follicular 21.00 ± 3.54 20.15 ± 3.60 −0.53a

Uncontrolled eatingLate-luteal 23.67 ± 5.23 19.54 ± 3.07 2.16*,a

Follicular 18.33 ± 3.84 17.92 ± 2.75 0.33a

Emotional eatingLate-luteal 8.00 ± 2.30 5.92 ± 1.32 2.35*,a

Follicular 7.83 ± 2.21 6.00 ± 1.78 2.23*,a

Leptin/BFP: serum leptin level divided by body fat percentage.a Non-parametric analysis: Z values of the Mann—Whitney Test.

ee1p

* p < 0.05.*** p < 0.001.

xhibit an increased basal metabolic rate and a greaternergy expenditure in the luteal phase (Solomon et al.,982). Thus, a higher luteal leptin level plays a role inreparing the body for the energy-intensive demands of

ppei

regnancy (Tessier et al., 2013). However, in line with arevious study (Akturk et al., 2013), the women with PMDDxhibited a decreased leptin level in the late-luteal phasen this study. This decline was significant when controlling

iitots

4b

TawsPte(Phh

cebrAmtIcagChohm

lwosobtrsbeTpfpt

4

Leptin in PMDD

for BFP and was significant only among the normal-weightwomen with PMDD. Owing to the moderating effect of beingoverweight on the association between PMDD and leptinlevel, the role of leptin in caloric intake could be differentin normal-weight and overweight women with PMDD.

4.4. The leptin level, caloric intake and eatingbehavior among the normal-weight women withPMDD

The normal-weight women with PMDD had a lower leptinlevel than the control group when controlling for BFP inthe late-luteal phase. The luteal peak in the leptin levelis important in terms of the luteal-follicular and the luteal-preimplantatory functional shifts (Geisthovel et al., 2004)and essential for fertility function (Goumenou et al., 2003).How or whether the late-luteal decline in the leptin levelamong women with PMDD alters the fertility function shouldbe evaluated in future study.

In the normal-weight subjects, the leptin level was nega-tively correlated with caloric intake in the late-luteal phase.Low levels of leptin have been identified as the strongestsignals to induce adaptive biological actions such as anincreased energy intake and a reduced energy expenditure(Zheng et al., 2009). As the women with PMDD had a lowerleptin level in the late-luteal phase, they had a higher caloricintake in this study. Further, leptin could inhibit sweet tasteresponse (Ninomiya et al., 2002). Thus, the sweet fooddrives a stronger appetite (Rasheed and Al-Sowielem, 2003)and a craving response (Yen et al., 2010) among women withPMDD in the late-luteal phase. In this study, the women withPMDD with a lower leptin/BFP had a higher late-luteal sweetcaloric intake, indicating a possible role of a low leptin levelin the increased sweet intake in the late-luteal phase.

Further, the normal-weight women with PMDD had higheruncontrolled eating with late-luteal exacerbation. Food isa natural reward to drive the appetite in normal subjects(Volkow et al., 2011). Uncontrolled eating occurs when theappetite overrides the control of food intake (Volkow et al.,2011) and contributes to sweet intake (Turner et al., 2010).As sweet food drives a positive emotional response in thelate-luteal phase (Yen et al., 2010), and women with PMDDhave a higher reward sensitivity (Ko et al., 2014) and greaterimpulsivity (Yen et al., 2011), they are vulnerable to uncon-trolled eating in the late-luteal phase.

Moreover, normal-weight women with PMDD have higheremotional eating, which is an eating behavior motivated bymultiple negative emotions (Chesler, 2012). Depression hasbeen found to be associated with the late-luteal increase inappetite of women with PMDD (Both-Orthman et al., 1988).A negative mood triggers overeating, and overeating mayfunction as a strategy to escape, avoid, or minimize nega-tive affect (Zysberg and Rubanov, 2010; Raman et al., 2013).However, the pleasure response is only temporary (Machtand Mueller, 2007) and does not provide a lasting benefitto the mood state. A persistent negative mood can even-tually increase the food intake (Willner, 1998). Our result

supported the late-luteal emotional eating pattern amongnormal-weight women with PMDD.

Thus, the lower leptin level, uncontrolled eating, andemotional eating combine to involve in the late-luteal

Nss

59

ncreased caloric intake, particularly the sweet caloricntake, among normal-weight women with PMDD. These fac-ors should be addressed in order to prevent late-lutealvereating among normal-weight women with PMDD, andhe effects of intervention should be evaluated in futuretudy.

.5. The leptin level, caloric intake and eatingehavior among the overweight women with PMDD

he difference in the leptin level between the overweightnd normal-weight subjects was more significant among theomen with PMDD than among the controls. This demon-

trated hyperleptinemia among the overweight women withMDD. The higher body lipid percentage might contributeo a higher leptin level in overweight subjects (Minoccit al., 2000). However, they had also a higher leptin/BFPZ = −4.11, p < 0.001) than the normal-weight women withMDD when the body lipid percentage was controlled. Thus,yperleptinemia could not be attributed simply to theirigher fat percentage.

Two major biological mechanisms have been defined thatontribute to leptin resistance under a high-fat diet. In thearly stage, leptin is transported across the blood-brain-arrier by a saturable process and saturation occurs atelatively low levels of circulating leptin (Banks et al., 1996).s adiposity and circulating leptin levels increase, impair-ent of BBB leptin transport increasingly fails to translate

he increases in serum leptin into increases in CNS leptin.n the later stage, elevated plasma leptin levels result inhronic overstimulation of leptin receptors. This results inlterations in leptin receptor expression and second messen-er signaling in the arcuate nucleus and other areas of theNS (Vasselli et al., 2013). These two models support theypothesis that hyperleptinemia contributes to the devel-pment of leptin resistance (Knight et al., 2010). Thus,yperleptinemia in overweight women with PMDD couldake them vulnerable to leptin resistance.Further, diet itself plays essential roles in obesity and

eptin resistance (Bray, 2010). The overweight womenith PMDD had a higher sweet caloric intake than theverweight controls. In Taiwan, heavy high-fructose cornyrup-containing beverages are popular, and a total of 25.1%f adolescents drink >500 ml per day of sugar-sweetenedeverages (Lin et al., 2013). A heavy fructose intake con-ributes not only to obesity (Bray, 2010), but also to leptinesistance (Vasselli et al., 2013). Thus, the higher late-lutealweet caloric intake in overweight women with PMDD shoulde assessed and addressed. Moreover, they also had highermotional eating behaviors than the overweight controls.hus, intervention for emotional symptoms, dietary control,articularly control of high-fructose corn syrup-containingood, and a healthy lifestyle, such as exercise, should beromoted for overweight women with PMDD to attenuatehe risk of leptin resistance.

.6. Limitations

onetheless, several limitations should be noted in thistudy. First, the number of subjects was limited in thistudy because many candidates were excluded according to

6

ttatatt

4

Titwwdctwoolcawoow

R

FCNvcif

C

T

A

TeK

R

A

A

A

A

B

B

B

B

C

C

C

D

E

E

F

F

G

G

G

H

J

0

he criteria for a symptomatic cycle (Smith et al., 2003) inhe prospective investigation. Second, the leptin level wasssessed during starvation, but not during satiation. Evalua-ion of the response of the leptin level to eating food mightllow more in-depth evaluation of the leptin level. Third,he causal relationships could not be confirmed owing tohe cross-sectional design of this study.

.7. Conclusion

he women with PMDD had greater increases in caloricntake, sweet caloric intake, and uncontrolled eating inhe late-luteal phase than the controls. Among the normal-eight women, the leptin level was negatively correlatedith caloric intake. The leptin level of the women with PMDDeclined in the late-luteal phase and was negatively asso-iated with sweet caloric intake. These results indicatedhat the late-luteal decline in the leptin level, integratedith uncontrolled eating and emotional eating, is involved invereating among normal-weight women with PMDD. On thether hand, the overweight women with PMDD had hyper-eptinemia in the late-luteal phase. The hyperleptinemiaould be a possible promotor of increased caloric intakemong overweight women with PMDD, as well as overweightomen without PMDD. The detailed roles of leptin andther psycho-neuro-endocrinology factors in the mechanismf eating behaviors among normal-weight or overweightomen with PMDD should be studied in the future.

ole of the funding source

unding for this study was provided by National Scienceouncil and the Kaohsiung Medical University Hospital; theational Science Council and the Kaohsiung Medical Uni-ersity Hospital had no further role in study design; in theollection, analysis and interpretation of data; in the writ-ng of the report; and in the decision to submit the paperor publication.

onflict of interest

he authors declare no conflict of interest.

cknowledgments

his study was supported by grants from the National Sci-nce Council Taiwan (NSC 100-2629-B-037-001-MY2) and theaohsiung Medical University Hospital (KMUH100-0R51).

eferences

hrens, K., Mumford, S.L., Schliep, K.C., Kissell, K.A., Perkins, N.J.,Wactawski-Wende, J., Schisterman, E.F., 2014. Serum leptin lev-els and reproductive function during the menstrual cycle. Am.J. Obstet. Gynecol. 210, 248.e241—248.e249.

kturk, M., Toruner, F., Aslan, S., Altinova, A.E., Cakir, N., Elbeg,S., Arslan, M., 2013. Circulating insulin and leptin in women withand without premenstrual dysphoric disorder in the menstrualcycle. Gynecol. Endocrinol. 29, 465—469.

C.-H. Ko et al.

merican Psychiatric Association, 2012. The draft fifth edition of theDiagnostic and Statistical Manual of Mental Disorders (DSM-5).American Psychiatric Association, Arlington.

merican Psychiatric Association, 2000. Diagnostic and StatisticalManual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR). American Psychiatric Association, Arlington.

anks, W.A., Kastin, A.J., Huang, W., Jaspan, J.B., Maness, L.M.,1996. Leptin enters the brain by a saturable system independentof insulin. Peptides 17, 305—311.

oth-Orthman, B., Rubinow, D.R., Hoban, M.C., Malley, J., Grover,G.N., 1988. Menstrual cycle phase-related changes in appetitein patients with premenstrual syndrome and in control subjects.Am. J. Psychiatry 145, 628—631.

ray, G.A., 2010. Soft drink consumption and obesity: it is all aboutfructose. Curr. Opin. Lipidol. 21, 51—57.

uffenstein, R., Poppitt, S.D., McDevitt, R.M., Prentice, A.M., 1995.Food intake and the menstrual cycle: a retrospective analy-sis, with implications for appetite research. Physiol. Behav. 58,1067—1077.

hesler, B.E., 2012. Emotional eating: a virtually untreated risk fac-tor for outcome following bariatric surgery. Sci. World J. 2012,365961.

hou, S.H., Chamberland, J.P., Liu, X., Matarese, G., Gao, C.,Stefanakis, R., Brinkoetter, M.T., Gong, H., Arampatzi, K.,Mantzoros, C.S., 2011. Leptin is an effective treatment forhypothalamic amenorrhea. Proc. Natl. Acad. Sci. U.S.A. 108,6585—6590.

ohen Jr., M.M., 2006. Role of leptin in regulating appetite, neu-roendocrine function, and bone remodeling. Am. J. Med. Genet.A 140, 515—524.

ye, L., Blundell, J.E., 1997. Menstrual cycle and appetite con-trol: implications for weight regulation. Hum. Reprod. 12,1142—1151.

pperson, C.N., Steiner, M., Hartlage, S.A., Eriksson, E., Schmidt,P.J., Jones, I., Yonkers, K.A., 2012. Premenstrual dysphoric dis-order: evidence for a new category for DSM-5. Am. J. Psychiatry169, 465—475.

vans, S.M., Foltin, R.W., Fischman, M.W., 1999. Food ‘‘cravings’’and the acute effects of alprazolam on food intake in womenwith premenstrual dysphoric disorder. Appetite 32, 331—349.

eng, H., Zheng, L., Feng, Z., Zhao, Y., Zhang, N., 2013. The roleof leptin in obesity and the potential for leptin replacementtherapy. Endocrine 44, 33—39.

enichel, R.M., Dominguez, J.E., Mayer, L., Walsh, B.T., Boozer, C.,Warren, M.P., 2008. Leptin levels and luteinizing hormone pul-satility in normal cycling women and their relationship to dailychanges in metabolic rate. Fertil. Steril. 90, 1161—1168.

eisthovel, F., Jochmann, N., Widjaja, A., Horn, R., Brabant, G.,2004. Serum pattern of circulating free leptin, bound leptin,and soluble leptin receptor in the physiological menstrual cycle.Fertil. Steril. 81, 398—402.

iannini, A.J., Price, W.A., Loiselle, R.H., Giannini, M.C., 1985.Hyperphagia in premenstrual tension syndrome. J. Clin. Psychi-atry 46, 436—438.

oumenou, A.G., Matalliotakis, I.M., Koumantakis, G.E., Panidis,D.K., 2003. The role of leptin in fertility. Eur. J. Obstet. Gynecol.Reprod. Biol. 106 (2), 118—124.

siao, M.C., Liu, C.Y., Hsu, S.C., Hsiao, C.C., Lin, Y.H., Hsieh, T.T.,2011. Elevated serum cholesterol levels in women with premen-strual dysphoric disorder. Int. J. Psychiatry Med. 42, 85—92.

auch-Chara, K., Oltmanns, K.M., 2014. Obesity — a neuropsy-chological disease? Systematic review and neuropsychologicalmodel. Prog. Neurobiol. 114, 84—101.

Karlsson, J., Persson, L.O., Sjostrom, L., Sullivan, M., 2000. Psy-chometric properties and factor structure of the Three-Factor

Eating Questionnaire (TFEQ) in obese men and women. Resultsfrom the Swedish Obese Subjects (SOS) study. Int. J. Obes. Relat.Metab. Disord. 24, 1715—1725.

S

S

S

S

S

T

T

v

V

V

V

v

W

Y

Y

Y

Z

Leptin in PMDD

Klok, M.D., Jakobsdottir, S., Drent, M.L., 2007. The role of leptinand ghrelin in the regulation of food intake and body weight inhumans: a review. Obes. Rev. 8, 21—34.

Knight, Z.A., Hannan, K.S., Greenberg, M.L., Friedman, J.M., 2010.Hyperleptinemia is required for the development of leptin resis-tance. PLoS One 5, e11376.

Ko, C.H., Long, C.Y., Chen, S.Y., Chen, I.J., Huang, T.H., Yen, J.Y.,2013. Depression, irritability, and anxiety in women with pre-menstrual dysphoric disorder. Int. J. Psychiatry Med. 46, 39—55.

Ko, C.H., Long, C.Y., Yen, C.F., Chen, C.S., Wang, P.W., Yen, J.Y.,2014. Gonadotrophic hormone and reinforcement sensitivity sys-tems in women with premenstrual dysphoric disorder. PsychiatryClin. Neurosci. 68, 785—794.

Kronick, I., Auerbach, R.P., Stich, C., Knauper, B., 2011. Compen-satory beliefs and intentions contribute to the prediction ofcaloric intake in dieters. Appetite 57, 435—438.

Leitch, M.A., Morgan, M.J., Yeomans, M.R., 2013. Different subtypesof impulsivity differentiate uncontrolled eating and dietaryrestraint. Appetite 69, 54—63.

Lin, W.T., Huang, H.L., Huang, M.C., Chan, T.F., Ciou, S.Y., Lee,C.Y., Chiu, Y.W., Duh, T.H., Lin, P.L., Wang, T.N., Liu, T.Y., Lee,C.H., 2013. Effects on uric acid, body mass index and bloodpressure in adolescents of consuming beverages sweetened withhigh-fructose corn syrup. Int. J. Obes. (Lond.) 37, 532—539.

Macht, M., Mueller, J., 2007. Immediate effects of chocolate onexperimentally induced mood states. Appetite 31, 667—674.

Masho, S.W., Adera, T., South-Paul, J., 2005. Obesity as a risk factorfor premenstrual syndrome. J. Psychosom. Obstet. Gynaecol.26, 33—39.

McNeil, J., Doucet, E., 2012. Possible factors for altered energybalance across the menstrual cycle: a closer look at the severityof PMS, reward driven behaviors and leptin variations. Eur. J.Obstet. Gynecol. Reprod. Biol. 163, 5—10.

Minocci, A., Savia, G., Lucantoni, R., Berselli, M.E., Tagliaferri, M.,Calò, G., Petroni, M.L., de Medici, C., Viberti, G.C., Liuzzi, A.,2000. Leptin plasma concentrations are dependent on body fatdistribution in obese patients. Int. J. Obes. Relat. Metab. Disord.24, 1139—1144.

Myers Jr., M.G., Heymsfield, S.B., Haft, C., Kahn, B.B., Laughlin, M.,Leibel, R.L., Tschöp, M.H., Yanovski, J.A., 2012. Challenges andopportunities of defining clinical leptin resistance. Cell Metab.15, 150—156.

Ninomiya, Y., Shigemura, N., Yasumatsu, K., Ohta, R., Sugimoto, K.,Nakashima, K., Lindemann, B., 2002. Leptin and sweet taste.Vitam. Horm. 64, 221—248.

Raman, J., Smith, E., Hay, P., 2013. The clinical obesity main-tenance model: an integration of psychological constructsincluding mood, emotional regulation, disordered overeating,habitual cluster behaviours, health literacy and cognitive func-tion. J. Obes. 2013, 240128.

Rapkin, A.J., Winer, S.A., 2009. Premenstrual syndrome and pre-menstrual dysphoric disorder: quality of life and burden ofillness. Expert Rev. Pharmacoecon. Outcomes Res. 9, 157—170.

Rasheed, P., Al-Sowielem, L.S., 2003. Prevalence and predictorsof premenstrual syndrome among college-aged women in SaudiArabia. Ann. Saudi Med. 23, 381—387.

Reed, S.C., Levin, F.R., Evans, S.M., 2008. Changes in mood, cogni-tive performance and appetite in the late luteal and follicularphases of the menstrual cycle in women with and without PMDD(premenstrual dysphoric disorder). Horm. Behav. 54, 185—193.

Z

61

chulz, C., Paulus, K., Johren, O., Lehnert, H., 2012. Intranasalleptin reduces appetite and induces weight loss in rats with diet-induced obesity (DIO). Endocrinology 153, 143—153.

mith, M.J., Schmidt, P.J., Rubinow, D.R., 2003. OperationalizingDSM-IV criteria for PMDD: selecting symptomatic and asymp-tomatic cycles for research. J. Psychiatr. Res. 37, 75—83.

olomon, S.J., Kurzer, M.S., Calloway, D.H., 1982. Menstrual cycleand basal metabolic rate in women. Am. J. Clin. Nutr. 36,611—616.

teiner, M., Macdougall, M., Brown, E., 2003. The PremenstrualSymptoms Screening Tool (PSST) for clinicians. Arch. WomensMent. Health 6, 203—209.

uzuki, K., Jayasena, C.N., Bloom, S.R., 2012. Obesity and appetitecontrol. Exp. Diabetes Res. 2012, 824305.

essier, D.R., Ferraro, Z.M., Gruslin, A., 2013. Role of leptin inpregnancy: consequences of maternal obesity. Placenta 34,205—211.

urner, S.A., Luszczynska, A., Warner, L., Schwarzer, R., 2010. Emo-tional and uncontrolled eating styles and chocolate chip cookieconsumption. A controlled trial of the effects of positive moodenhancement. Appetite 54, 143—149.

an Strien, T., Ouwens, M.A., Engel, C., de Weerth, C., 2014.Hunger, inhibitory control and distress-induced emotional eat-ing. Appetite 79, 124—133.

an Vugt, D.A., 2010. Brain imaging studies of appetite in the con-text of obesity and the menstrual cycle. Hum. Reprod. Update16, 276—292.

asselli, J.R., Scarpace, P.J., Harris, R.B., Banks, W.A., 2013.Dietary components in the development of leptin resistance.Adv. Nutr. 4, 164—175.

olkow, N.D., Wang, G.J., Baler, R.D., 2011. Reward, dopamine andthe control of food intake: implications for obesity. Trends Cogn.Sci. 15, 37—46.

on Schnurbein, J., Moss, A., Nagel, S.A., Muehleder, H., Debatin,K.M., Farooqi, I.S., Wabitsch, M., 2012. Leptin substitutionresults in the induction of menstrual cycles in an adolescent withleptin deficiency and hypogonadotropic hypogonadism. Horm.Res. Paediatr. 77, 127—133.

illner, J.P., 1998. Reproductive genetics and today’spatient options: prenatal diagnosis. Mt. Sinai J. Med. 65,173—177.

en, J.Y., Chang, S.J., Ko, C.H., Yen, C.F., Chen, C.S., Yeh,Y.C., Chen, C.C., 2010. The high-sweet-fat food cravingamong women with premenstrual dysphoric disorder: emotionalresponse, implicit attitude and rewards sensitivity. Psychoneu-roendocrinology 35, 1203—1212.

en, J.Y., Chen, C.C., Chang, S.J., Ko, C.H., Chen, C.S., Yen,C.F., 2011. Hostility, impulsivity, and behavior inhibition amongwomen with PMDD. CNS Spectr. 16, 205—213.

en, J.Y., Tu, H.P., Chen, C.S., Yen, C.F., Long, C.Y., Ko, C.H., 2013.The effect of serotonin 1A receptor polymorphism on the cog-nitive function of premenstrual dysphoric disorder. Eur. Arch.Psychiatry Clin. Neurosci. 264, 729—739.

heng, H., Lenard, N.R., Shin, A.C., Berthoud, H.R., 2009. Appetitecontrol and energy balance regulation in the modern world:reward-driven brain overrides repletion signals. Int. J. Obes.

(Lond.) 33 (Suppl. 2), S8—S13.

ysberg, L., Rubanov, A., 2010. Emotional intelligence and emo-tional eating patterns: a new insight into the antecedents ofeating disorders? J. Nutr. Educ. Behav. 42, 345—348.