Universidade de São Paulo
2012
The effect of overweight/obesity on cognitive
function in euthymic individuals with bipolar
disorder
EUROPEAN PSYCHIATRY, PARIS, v. 27, n. 3, p. 223-228, APR, 2012
http://www.producao.usp.br/handle/BDPI/42806
Downloaded from: Biblioteca Digital da Produção Intelectual - BDPI, Universidade de São Paulo
Biblioteca Digital da Produção Intelectual - BDPI
Original
article
The
effect
of
overweight/obesity
on
cognitive
function
in
euthymic
individuals
with
bipolar
disorder
C.Y.
Yim
a,b,
J.K.
Soczynska
b,c,
S.H.
Kennedy
b,c,d,
H.O.
Woldeyohannes
b,
E.
Brietzke
e,
R.S.
McIntyre
a,b,*
,c,da
DepartmentofPharmacologyandToxicology,UniversityofToronto,Toronto,ON,Canada
bPsychiatryandPharmacology,MoodDisordersPsychopharmacologyUnit,UniversityHealthNetwork,399BathurstStreet,Toronto,ON,M5T2S8Canada cInstituteofMedicalScience,UniversityofToronto,Toronto,ON,Canada
d
DepartmentofPsychiatry,UniversityofToronto,Toronto,ON,Canada
e
BipolarDisorderProgram,InstituteofPsychiatry,UniversityofSa˜oPaulo,Sa˜oPaulo,Brazil
1. Introduction
Bipolardisorder(BD)hasbeenhighlyassociatedwithdisparate cognitivedeficitsincluding attention,psychomotorperformance, executive function,verbal fluency, learning, memory and global
neurocognitive functioning [2,25–27]. Available evidence also
indicatesthatindividualswithBDexhibitcognitivedeficitsnotonly duringacutemoodepisodesbutalsoduringeuthymia[26,27,41,43]. Moreover, a recent meta-analysis reported that euthymic individuals generallyhadthesamelevelofdeficitsinmemoryandlearningasan
activelysymptomaticgroup[18].Emergingevidencealsoindicates that first-degreerelativesofprobands withBD exhibita similar patternofcognitivedeficits,providingthebasesforhypothesizing
that cognitive abnormalities may represent an endophenotypic
markerofBD[4,43].ThepertinacityofcognitivedeficitsinBDis
underscored by reports documenting an association between
cognitivedeficitsandpsychosocialfunctioning,workforce perfor-manceandinterpersonaladjustment[17,32,47].
Emergingevidencealsoindicatesthatobesityisassociatedwith
reduced cognitive function in otherwise healthy individuals
[6,10,11,14,16,19,23,39].Theassociationbetweenanthropometrics andcognitivedeficitsisdetectableinindividualswithout
obesity-associated co-morbidities (e.g. type 2 diabetes mellitus and
hypertension) known to independently affect brain function
ARTICLE INFO Articlehistory:
Received22October2010
Receivedinrevisedform14January2011 Accepted6February2011
Availableonline12May2011
Keywords: Overweight Obesity Cognitivefunction Euthymic Bipolar ABSTRACT
Background. – Persistentimpairmentincognitivefunctionhasbeendescribedineuthymicindividuals withbipolar disorder.Collectivework indicates thatobesityis associated withreducedcognitive functioninotherwisehealthyindividuals.Thissub-grouppost-hocanalysispreliminarilyexploresand examinestheassociationbetweenoverweight/obesityandcognitivefunctionineuthymicindividuals withbipolardisorder.
Methods.– EuthymicadultswithDSM-IV-TR-definedbipolarIorIIdisorderwereenrolled.Subjects includedinthispost-hocanalysis(n=67)weredividedintotwogroups(normalweight,bodymass index [BMI] of 18.5–24.9kg/m2; overweight/obese, BMI25.0kg/m2). Demographic and clinical
informationwereobtainedatscreening.Atbaseline,studyparticipantscompletedacomprehensive cognitivebattery toassesspremorbidIQ,verballearningandmemory,attentionand psychomotor processingspeed,executivefunction,generalintellectualabilities,recollectionandhabitmemory,as wellasself-perceptionsofcognitivefailures.
Results. –BMIwasnegativelycorrelatedwithattentionandpsychomotorprocessingspeedasmeasured bytheDigit SymbolSubstitution Test (P<0.01). Overweightand obese bipolarindividuals had a significantly lowerscore on the VerbalFluency Test when compared to normal weight subjects (P<0.05).Forallothermeasuresofcognitivefunction,non-significanttrendssuggestinganegative associationwithBMIwereobserved,withtheexceptionofmeasuresofexecutivefunction(i.e.Trail MakingTestB)andrecollectionmemory(i.e.process-dissociationtask).
Conclusion. –Notwithstandingthepost-hocmethodologyandrelativelysmallsamplesize,theresults ofthisstudysuggestapossiblenegativeeffectofoverweight/obesityoncognitivefunctionineuthymic individualswithbipolardisorder.Takentogether,thesedataprovidetheimpetusformorerigorous evaluationofthemediationalroleofoverweight/obesity(andothermedicalco-morbidity)oncognitive functioninpsychiatricpopulations.
ß2011ElsevierMassonSAS.Allrightsreserved.
*Correspondingauthor.Tel.:+4166035279;fax:+4166035368. E-mailaddress:[email protected](R.S.McIntyre).
0924-9338/$–seefrontmatterß2011ElsevierMassonSAS.Allrightsreserved. doi:10.1016/j.eurpsy.2011.02.004
[19,20,39].Mostcognitivedomainsarereportedtobeadversely affectedbyexcessweightwithreplicatedabnormalitiesinmeasures oflearning,memoryandexecutivefunction[10].Abi-directional relationshipbetweenobesityandcognitivefunctionissuggestedby studiesreporting thatindividuals withimpairment in executive functionaremorelikelytobecomeoverweightorobese,perhaps relatedto disturbances in impulse control, self-monitoring, and
goal-directed behaviour [24]. It is further hypothesized that
cognitiveabnormalitiesobservedinoverweight/obeseindividuals aretheexpressionofabnormalitiesinbrainstructureandfunction [11,16,36].
Ithasbeenamplydocumentedthat individualswithBD are
differentiallyassociatedwithoverweight/obesityandabdominal obesity,andexcessweightadverselyeffectsillnesspresentation,
courseand outcome; however,toourknowledge,no published
study has primarily examined the association between
over-weight/obesity and cognitive function in adults with BD [15].
Herein, wepreliminarily soughttodetermine whether suchan
association exists in a well-characterized clinical cohort of
euthymicadultswithBDI/II.
2. Methods
2.1. Overview
Thedataforthepresentanalysiswereobtainedpost-hocfrom screeningandbaselineassessmentsofeuthymicindividualswith BDenrolledinastudythatprimarilyaimstoevaluateintranasal insulinasatherapeuticinterventionforcognitivefunction.Forthe purposeoftheanalysisherein,participantswerecategorizedinto
normal weightand overweight/obesebased on theestablished
criteriaofbody indexmass(BMI)(normalweight,BMIof18.5– 24.9kg/m2;overweight/obese,BMI25.0kg/m2)[37].
2.2. Participants
Eligibleparticipantsfortheprincipalstudy(age:18–60)were euthymicindividualswithDSM-IV-TR-definedBDI/II,confirmedby the MiniInternationalNeuropsychiatricInterview (MINI) [1,42]. Euthymia,definedasascoreoflessorequalto3on the7-item HamiltonRatingScaleforDepression(HAMD-7)andascoreofless orequalto7ontheYoungManiaRatingScale(YMRS)wasconfirmed attheinitialscreeningand1monthlateratbaseline[33,43,48,49].
Subjectswere permittedtomaintain their medicationregimens
suchasconventionalBDpharmacotherapy.Nochangesin medica-tionwereallowedduringthestudy;medicationswerecontinuedat thetimeofcognitiveassessment.Otherinclusioncriteriaincluded
good physical health verified by a physical exam, provision of
informed consent and use of a medically accepted means of
contraception for females. Exclusion criteria included other
concurrentDSM-IV-definedAxis-Idiagnosesclinicallysignificant untreated medical conditions (e.g. cardiovascular, neurological, gastrointestinal,hematological,renal,hepatic,respiratoryor
endo-crine illnesses), uncorrected hypo/hyperthyroidism (including
elevatedthyroidstimulating hormone),the presence ofdiabetes mellitus or hypo/hyperglycemia, history of neurological trauma resultinginlossofconsciousness,currentpregnancyor breastfeed-ing,orhistoryofpregnancyinthelast12months,electroconvulsive therapyin the preceding6months,substance oralcohol abuse/ dependenceinthelast3months(meetingDSM-IVcriteria)andBMI greaterorequalto40kg/m2[1].
2.3. Procedures
Participants were largely recruited from referrals to the
outpatientMoodDisordersPsychopharmacologyUnit,University
HealthNetwork,UniversityofTorontoandfrommedia announce-mentsatlocalhospitalsinthecommunity. Theireligibilitywas determinedattheinitialscreeningvisitwhereintheirclinicalstate
(i.e. euthymia) and type of BD were confirmed. A physical
examinationwasconductedtoexcludethosewithmajormedical
conditions. After 2weeks, eligible participants who provided
writteninformedconsentand completedtheinitialassessment
wereinvited tothesecond screeningduring whichtheir blood
pressure,pulse,height,weight,waist-to-hip ratioandBMIwere measuredandtheirbloodandurinesampleswerecollectedaspart
of the evaluation. Demographic information and psychiatric
history wereobtained bydirect interview.Priortothebaseline
visit (which took place 4weeks after the initial screening),
participants were asked to abstain from alcohol and smoking
cigarettesfor48hoursand30minutes,respectively,priortothe
blood sample collection. At baseline, euthymia wasre-verified
prior to cognitive testing administered by trained research
personnel.
2.4. Cognitivemeasures
A battery of cognitive tests was administered to assess
estimatedpremorbidIQ(NationalAdultReadingTest[NART-R]
[38]),verballearning andmemory (California Verbal Learning
Test[CVLT-II][12]),attentionandpsychomotorprocessingspeed (TrailMakingTestA[TMT-A][40]andDigitSymbolSubstitution Test[DSST][46]),executivefunction(TrailMakingTestB[TMT-B] [40]andVerbal FluencyTest [3]),generalintellectual abilities
(Shipley Abstraction [50]), recollection and habit memory
(Process-DissociationTask[21]),andself-perceptionsofcognitive failures(CognitiveFailuresQuestionnaire[CFQ])[5].The
cogni-tive measures that were chosen in this study were based on
the primary hypothesis around the efficacy of a cognitive
enhancingintervention.Thecognitivebatterywasadministered byatrainedresearchcoordinatorandtookapproximately3hours toadminister.
2.5. Statisticalanalyses
StatisticalanalyseswereperformedusingSPSSversion16.0 (SPSS Inc., Chicago, IL). Since the variables presented normal
distribution, parametric tests were chosen. Demographic and
clinical characteristics of normal weight subjects and
over-weight/obesesubjectswerecomparedusingChi-squaretestsfor
categorical variables and independent samples t-tests for
continuous variables.Correlations between BMI andmeasures
of cognitive performance were generated using Pearson’s
correlationtests.Performanceoncognitivetestswascompared
between the two groups using independent samples t-tests.
StatisticalsignificancewassetatP<0.05.Inordertoexamine the main effect of obesity oncognitive function, multivariate
analyses of covariance (MANCOVA’s) were conducted on test
performance in three different cognitive domains: verbal
learning and memory (measured by CVLT), attention and
psychomotorprocessingspeed(measuredbyTMT-AandDSST),
andexecutivefunction(measuredbyTMT-BandVerbalFluency
Test).The independentvariable in eachanalysisincludedBMI
groups(normalvs.overweight/obese).Variablessuchasageand
thenumber ofyears ofeducation countedfrom Grade1 were
includedineachmodelascovariates,whichshowedasignificant
association with dependent variables based on Pearson’s
correlation tests (P<0.05). Binary logistic regression was
performedtoassess theassociationbetween thedichotomous
dependentvariable,BMIgroup,andpredictorvariablessuchas
age andthetotal number of correctresponses onthe Shipley
Abstractionsubtest.
C.Y.Yimetal./EuropeanPsychiatry27(2012)223–228 224
3. Results
3.1. Samplecharacteristics
Atotalof67euthymicindividualswithBDwereincludedinthe analysis.71.6%(n=48)ofthesamplewereoverweightorobese. DemographicandclinicalcharacteristicsarepresentedinTable1 (thesubjectsremoveddidnot differin demographicor clinical
parameters from the 60 who were analyzed). There were no
significant differences between normal and overweight/obese
groupsindemography,exceptforsexwhereinahigherproportion
of males were observed in the overweight/obese group than
expected.Thetwogroupsdidnotdifferinclinicalcharacteristics
including illness severity, age at the first mood episode, the
number of lifetimemood episodesand the number of lifetime
hospitalizationsfordepression/mania.Moreover,currentsmoking
status and family history of mental illness did not differ
significantly between the two groups (data not shown). All
subjects received conventional pharmacological treatment for
bipolardisorder.There wereno significantdifferencesbetween groupsinthenumberortypeofmedications.Metabolicsyndrome componentswereexclusion criteria,and assuch, therewereno differencesin lipidor cholesterolprofiles ormarkersof insulin resistance.Also,theresultsremainedunchangedafteradjusting fortheeffectsofgender.
3.2. CorrelationbetweenBMIandmeasuresofcognitivetest performance
BMIwassignificantlyandnegativelycorrelatedwithscoreson theDigitSymbolSubstitutionTest(r= 0.320,P<0.01).Formost oftheanalyses,atrendwasobservedinthehypothesizeddirection
wherein BMI was negatively correlated with various cognitive
measures.
3.3. Performanceoncognitivetestsbynormalweightandoverweight/ obeseindividualswithBD
Groupmean performanceand statistical comparisonsfor all cognitivemeasuresaresummarizedinTable2.Subjectswhowere overweightorobesehadasignificantlylowerscoreontheVerbal
Fluency Test compared to those who were normal weight
(P=0.013).Nosignificantdifferencesinperformanceonanyother cognitivemeasuresweredetectedbetweenthetwogroups. 3.4. Maineffectofobesityoncognitivedomains
Table 3 presents the main effect of obesity on cognitive
testperformance.ResultsfromMANCOVAanalysesdidnotyield
any significant main effect of obesity on cognitive function
domains(verballearningandmemory;attentionand
psychomo-Table1
Demographicandclinicalcharacteristicsofnormalweightandoverweight/obeseeuthymicindividualswithbipolardisorder(BD).
Characteristic Normalweight(n=19) Overweight/Obeseweight(n=48) Pvalue
Sex(n,%) 0.008c Female 14(73.7) 18(37.5) Male 5(26.3) 30(62.5) Age(mean,SD) 36.68(10.36) 41.48(9.90) 0.082 Ethnicity(n,%) 0.099 White 17(89.5) 45(93.8) Black 1(5.3) 0(0) Asian 0(0) 3(6.2) Other 1(5.3) 0(0)
NumberofyearsofeducationcountedfromGrade1(mean,SD) (N:16,O:44)a 16.78(2.61) 15.73(3.00) 0.220 TypeofBD(n,%) 0.776 BDI 16(84.2) 39(81.2) BDII 3(15.8) 9(18.8) Currentmedication(n)b Antidepressants 11 20 0.230 Sleepmedication 5 11 0.769 Anxiolytic 5 13 0.949 Anticonvulsants 15 38 0.984 Antipsychotic 11 30 0.727 Hormonal 5 8 0.368
Ageatfirstdepressiveepisode(mean,SD) (N=18,O=42)a
17.22(5.91) 19.50(9.78) 0.363
Ageatfirstmanicepisode(mean,SD) (N=17,O=38)a
22.65(6.10) 24.08(10.72) 0.610
Numberoflifetimedepressiveepisodes(n,SD) (N=13,O=28)a
6.23(8.10) 10.32(8.65) 0.159
Numberoflifetimemanicepisodes(n,SD) (N=13,O=28)a
6.46(9.56) 7.00(9.32) 0.865
Numberoflifetimehospitalizationsfordepression(n,SD) (N=18,O=42)a
0.72(1.13) 1.19(2.02) 0.360
Numberoflifetimehospitalizationsformania(n,SD) (N=17,O=43)a
0.71(0.92) 0.81(1.22) 0.743
a
N:thenumberofsubjectsinnormalweightgroup;O:thenumberofsubjectsinoverweight/obesegroup.
b
Themajorityofsubjectsreceivedpolypharmacy.
c
tor speed; and executive function) after accounting for the confounders.
3.5. PredictorofobesityineuthymicindividualswithBD
The sample size included in theregression analysis was 60 subjects(outliersand missingdata excludedfromtheanalysis) (Table4).Statisticalanalysesdemonstratedthatbothageandthe
score of the Shipley Abstraction subtest were not significant
predictorsofobesityineuthymicindividualswithBD. 4. Discussion
To ourknowledge,the analysis herein isthe firstattempt to assesswhetheroverweight/obesitywasassociatedwithdecrements
in cognitive function in euthymic individuals with BD. A high
proportionoftheparticipantsinthisstudywereoverweightorobese (71.6%), as has been previously reportedin bipolar populations [31,34,35].Possiblefactorscontributing toobesityinBD include lifestyle,medicationexposure,neuroendocrineand neurotransmit-terdysfunctions,geneticpredispositionandco-morbidconditions suchasbinge-eatingdisorder[30,31,34,35].Ourdatawereunableto determinewhetherobesityoccurredpriortotheonsetofBDorasa consequenceofmulti-episode/chronicillnessandtreatment.
In keeping with the view that increased body weight is
negatively correlatedwithcognitivefunctionin non-psychiatric samples,atrendwasobservedwhereinBMIwasinverselyrelated
tocognitive test performance in euthymic individualswithBD
[19].Moreover,inaseparatestudyevaluatingeuthymic individu-als,BMIwasnegativelycorrelatedwiththeGlobalAssessmentof
Table2
Performanceoncognitivemeasuresbynormalweightandoverweight/obeseeuthymicindividualswithBD.
Cognitivedomains Normalweight(n=19) Overweight/obese(n=48) Pvalue
Cognitivemeasures Mean(SD) Mean(SD)
PremorbidIQ 0.672
NARTestimatedfullscaleIQ 113.10(9.23) 112.14(8.08) Verballearningandmemory
CVLT
Trials1–5freerecalltotalcorrect 56.74(11.48) 54.33(12.05) 0.459
Short-delayfreerecall 11.16(3.70) 10.56(3.68) 0.553
Short-delaycuedrecall 12.32(3.15) 11.86(2.82) 0.543
Long-delayfreerecall 11.79(3.54) 11.43(3.46) 0.702
Long-delaycuedrecall 12.26(3.14) 12.02(3.10) 0.776
Attentionandpsychomotorprocessingspeed
TrailMakingTestA 32.12(14.76) 32.94(10.10) 0.798
(N=18,O=47)a
DigitSymbolSubstitutionTest 58.68(11.80) 53.38(11.26) 0.091
Executivefunction
TrailMakingTestB 72.56(26.89) 74.21(35.42) 0.859
(N=18,O=47)a
VerbalFluencyTest 70.06(19.50) 59.29(13.87) 0.013b
Generalintellectualabilities
ShipleyAbstractionsubtest 14.11(3.23) 13.90(4.01) 0.840
Memorydeficits HABCON Recollectionmemory 0.36(0.12) 0.34(0.18) 0.778 (N=14,O=33)a Habitmemory 0.57(0.11) 0.59(0.11) 0.515 (N=13,O=33)a Subjectivemeasure
CognitiveFailuresQuestionnaire 54.74(15.20) 57.46(16.45) 0.535
aN:thenumberofsubjectsinnormalweightgroup;O:thenumberofsubjectsinoverweight/obesegroup. b
Statisticallysignificant(P<0.05).
Table3
Maineffectofobesityoncognitivedomain.
Cognitivedomain(cognitivemeasuresused) Numberofsubjectsincluded inthemodel(n)
df Pillai’sF Pvalue
Verballearningandmemory(CVLT) 59 3,53 0.399 0.754
Attentionandpsychomotorprocessingspeed(TMT-A,DSST) 58 2,53 0.548 0.582
Executivefunction(TMT-B,VerbalFluencyTest) 57 2,52 1.991 0.393
Table4
Binarylogisticregressionmodelofindependentvariablespredictingobesityinsubjects(n=60).
Predictor OddsRatio(OR) 95%CIforOR Pvalue
Age 1.045 0.986–1.107 0.136
TotalcorrectresponsesonShipleyAbstraction 1.022 0.871–1.199 0.788
C.Y.Yimetal./EuropeanPsychiatry27(2012)223–228 226
Functioning(GAF)[8](theGAFhasbeenpositivelycorrelatedwith cognitivetestperformanceinseparatestudies[27,28]).Basedon thesepreviousfindings,itmaybeinferredthatBMIcontributesto decreasedcognitivefunctioningineuthymicindividualswithBD. Itiswellestablishedthatimpairmentsinexecutivefunctionare apparentinmixedpopulationsofindividualswithBDaswellas
obese individuals without psychiatric disorders [7,10,16,19,
22,26,27,28,41,45].ItcouldbehypothesizedthatobesityandBD areassociatedwithcommoncentralnervoussystemstructuraland/ or functional changes in brain regions that subserve cognitive functioning. For example, frontal cortical regions that mediate executive function are hypometabolic in depressed individuals; similarly,overweight/obeseindividualsmanifestreducedmetabolic activity, as well as atrophy, in several cortical and subcortical structures [13,41,44]. Moreover, the interrelationship between
obesityandmooddisordersmaybeduetoa pathophysiological
nexus that includes abnormalities in
hypothalamus-pituitary-adrenal axis function, inflammatory and metabolic systems,
disruptionofbraincircuitry,allofwhicharepotentialmediators ofcognitivefunction[9,10,28,43].Furtherstructuralandfunctional
investigations, as well as the establishment of mechanisms
mediatingcognitivedeficitsinobesityandBD,arerequiredbefore afirmconclusioncanbedrawn.
Several methodological deficiencies limit inferences and
interpretationsthatcanbemadefromthesedata.First,thisstudy wasasub-grouppost-hocanalysisthatwasnotdesignedapriorito
address the association of cognition with overweight/obesity.
Secondly, the sample size was relatively small, increasing the possibilityoftypeIIerror.Thirdly,thesamplewascomprisedof individualsenrolledinaclinicaltrialatauniversity-basedhealth sciencecentre,limitinggeneralizabilitytootherpatient popula-tions.Fourthly,therewasnoattemptto‘‘enrich’’thesamplefor individualswhohavecognitivedeficits.Asaresult,considerable heterogeneityincognitiveprofileswasobserved,whichmayhave also decreased the likelihood of detecting an association [22]
Fifthly, we did not have detailed information regarding the
durationofoverweight/obesity.Studiesinnon-psychiatric popu-lationsindicatethattheadverseeffectofoverweight/obesitymay beduration-dependent[19].Theinclusionofindividualswithpast
psychiatric co-morbidity was permitted, as was psychiatric
medicine,introducingadditional confounds[28,29].Finally, the obeseindividualsinthisstudydidnothaveco-morbid hyperten-sion,diabetesmellitusorcardiovasculardisease.Itisnotknownif
our study results are representative of most other obese
individualswhohaveobesity-associatedco-morbidity.
Taken together, this preliminary analysis is intended to be hypothesis-generatingratherthanconfirming.Itisourobjectiveto providetheimpetusformorerefinedevaluationofthe mediation-al/moderationalroleofoverweight/obesityoncognitivefunction in bipolar (and other psychiatric) populations. Hitherto,
over-weight/obesity and other medical co-morbidity pertinent to
cognitivefunction(e.g.diabetesmellitus)havenotbeen system-atically evaluated as variables possibly pertinent to cognitive function in psychiatric populations. Future studies should also considerwhetherweightlossstrategiesinBDofferasalutaryeffect on cognitive functionin additionto otherbenefits on physical health.Moreover,althoughwedidnotfindadifferencebetween bipolarIandbipolarIIpopulations,asufficientlypoweredstudy shouldattempttoenrollbothbipolarIandbipolarIIpopulations, ascognitivedeficitsareprominentinbothgroups.
Disclosureofinterest
ThiswassupportedbyagrantfromStanleyMedicalResearch
Institute and Bristol-Myers-Squibb Canada Inc. to Dr Roger
S.McIntyre.
References
[1]American Psychiatric Association, American Psychiatric Association,Task ForceonDSM-IV. Diagnosticand statistical manualofmental disorders. DSM-IV-TR.4thed.,textrevisioned.Washington,DC:AmericanPsychiatric Association;2000.
[2]BeardenCE,HoffmanKM,CannonTD.Theneuropsychologyand neuroanato-my of bipolar affective disorder: a critical review. Bipolar Disord 2001;3(3):106–50.
[3]BentonAL,HamsherK.Multilingualaphasiaexamination.IowaCity: Univer-sityofIowa;1976.
[4]BoraE,YucelM,PantelisC.Cognitiveendophenotypesofbipolardisorder:a meta-analysisofneuropsychologicaldeficitsineuthymicpatientsandtheir first-degreerelatives.JAffectDisord2009;113(1–2):1–20.
[5]Broadbent DE,CooperPF,FitzGeraldP,ParkesKR.TheCognitiveFailures Questionnaire(CFQ)anditscorrelates.BrJClinPsychol1982;21(Pt1):1–16. [6]BrookJS,ZhangC,SaarNS,BrookDW.Psychosocialpredictors,higherbody massindex,andaspectsofneurocognitivedysfunction.PerceptMotSkills 2009;108(1):181–95.
[7]Bruce-KellerAJ,KellerJN,MorrisonCD.ObesityandvulnerabilityoftheCNS. BiochimBiophysActa2009;1792(5):395–400.
[8]CalkinC,vanDV,RuzickovaM,SlaneyC,GarnhamJ,HajekT,etal.Canbody massindexhelppredictoutcomeinpatientswithbipolardisorder?Bipolar Disord2009;11(6):650–6.
[9]CapuronL,MillerAH.Cytokinesandpsychopathology:lessonsfrom interfer-on-alpha.BiolPsychiatry2004;56(11):819–24.
[10]CohenRA.Obesity-associatedcognitivedecline:excessweightaffectsmore thanthewaistline.Neuroepidemiology2010;34(4):230–1.
[11]CournotM,MarquieJC,AnsiauD,MartinaudC,FondsH,FerrieresJ,etal. Relationbetweenbodymassindexandcognitivefunctioninhealthy middle-agedmenandwomen.Neurology2006;67(7):1208–14.
[12]DelisDC,KramerJH,KaplanE,OberBA.CaliforniaVerbalLearningTest: CVLT-II.NewYork,NY:PsychologicalCorporation;2000.
[13]DrevetsWC,PriceJL,FureyML.Brainstructuralandfunctionalabnormalities inmooddisorders:implicationsforneurocircuitrymodelsofdepression.Brain StructFunct2008;213(1–2):93–118.
[14]EliasMF,EliasPK,SullivanLM,WolfPA,D’AgostinoRB.Obesity,diabetesand cognitive deficit: The Framingham Heart Study. Neurobiol Aging 2005;26(Suppl.1):11–6.
[15]FagioliniA,FrankE,HouckPR,MallingerAG,SwartzHA,BuysseDJ,etal. Prevalenceofobesityandweightchangeduringtreatmentinpatientswith bipolarIdisorder.JClinPsychiatry2002;63(6):528–33.
[16]FergenbaumJH,BruceS,LouW,HanleyAJ,GreenwoodC,YoungTK.Obesity andloweredcognitiveperformanceinaCanadianFirstNationspopulation. Obesity(SilverSpring)2009;17(10):1957–63.
[17]FriedmanES.Cognitivedeficitsinbipolarindividuals:implicationsforclinical practice.CNSNeurosciTher2009;15(4):305–6.
[18]GoodwinFK,JamisonKR,GhaemiSN.In:Manic-depressiveillness.Bipolar disordersandrecurrentdepression2nded,NewYork:OxfordUniversityPress; 2007.
[19]GunstadJ,PaulRH,CohenRA,TateDF,SpitznagelMB,GordonE.Elevatedbody massindexisassociatedwithexecutivedysfunctioninotherwisehealthy adults.ComprPsychiatry2007;48(1):57–61.
[20]GunstadJ,SpitznagelMB,PaulRH,CohenRA,KohnM,LuysterFS,etal.Body massindexandneuropsychologicalfunctioninhealthychildrenand adoles-cents.Appetite2008;50(2–3):246–51.
[21]JacobyLL.Invarianceinautomaticinfluencesofmemory:towardauser’s guidefortheprocess-dissociationprocedure.JExpPsycholLearnMemCogn 1998;24(1):3–26.
[22]JamrozinskiK.Doeuthymicbipolarpatientshavenormalcognitive function-ing?CurrOpinPsychiatry2010;23(3):255–60.
[23]JeongSK,NamHS,SonMH,SonEJ,ChoKH.Interactiveeffectofobesityindexes oncognition.DementGeriatrCognDisord2005;19(2–3):91–6.
[24]LezakM,HowiesonD,LoringD.In:Neuropsychologicalassessment4thed,New York:OxfordUniversityPress;2004.
[25]MalhiGS, IvanovskiB,Hadzi-PavlovicD, MitchellPB,VietaE,SachdevP. Neuropsychologicaldeficitsandfunctionalimpairmentinbipolardepression, hypomaniaandeuthymia.BipolarDisord2007;9(1–2):114–25.
[26]Martinez-AranA,VietaE,ColomF,ReinaresM,BenabarreA,GastoC,etal. Cognitivedysfunctionsinbipolardisorder:evidenceofneuropsychological disturbances.PsychotherPsychosom2000;69(1):2–18.
[27]Martinez-AranA,VietaE,ReinaresM,ColomF,TorrentC,Sanchez-MorenoJ, etal.Cognitivefunctionacrossmanicorhypomanic,depressed,andeuthymic statesinbipolardisorder.AmJPsychiatry2004;161(2):262–70.
[28]Martinez-AranA,VietaE,ColomF,TorrentC,Sanchez-MorenoJ,ReinaresM, etal.Cognitiveimpairmentineuthymicbipolarpatients:implicationsfor clinicalandfunctionaloutcome.BipolarDisord2004;6(3):224–32. [29]Martinez-AranA,VietaE,TorrentC,Sanchez-MorenoJ,GoikoleaJM,Salamero
M,etal.Functionaloutcomeinbipolardisorder:theroleofclinicaland cognitivefactors.BipolarDisord2007;9(1–2):103–13.
[30]McElroySL,FryeMA,HellemannG,AltshulerL,LeverichGS,SuppesT,etal. Prevalenceandcorrelatesofeatingdisordersin875patientswithbipolar disorder.JAffectDisord2010;128(3):191–8.
[31]McElroySL,FryeMA,SuppesT,DhavaleD,KeckJrPE,LeverichGS,etal. Correlatesofoverweightandobesityin644patientswithbipolardisorder.J ClinPsychiatry2002;63(3):207–13.
[32]McIntyreRS,KonarskiJZ.Bipolardisorder:anationalhealthconcern.CNS Spectr2004;9(11Suppl.12):6–15.
[33]McIntyreRS,KonarskiJZ,ManciniDA,FultonKA,ParikhSV,GrigoriadisS,etal. Measuringtheseverityofdepressionandremissioninprimarycare: valida-tionoftheHAMD-7scale.CMAJ2005;173(11):1327–34.
[34]McIntyreRS,KonarskiJZ,WilkinsK,SoczynskaJK,KennedySH.Obesityin bipolar disorderand major depressive disorder:results from anational communityhealthsurveyonmentalhealthandwell-being.CanJPsychiatry 2006;51(5):274–80.
[35]McIntyreRS,NguyenHT,SoczynskaJK,LourencoMT,WoldeyohannesHO, KonarskiJZ.Medicalandsubstance-relatedco-morbidityinbipolardisorder: translationalresearchandtreatmentopportunities.DialoguesClinNeurosci 2008;10(2):203–13.
[36]McIntyreRS,KennaHA,NguyenHT,LawCW,SultanF,WoldeyohannesHO, etal.Brainvolumeabnormalities andneurocognitivedeficitsindiabetes mellitus:pointsofpathophysiologicalcommonalitywithmooddisorders? AdvTher2010;27(2):63–80.
[37]NationalHeartLaBI,NationalInstituteofDiabetesandDigestiveandKidney Diseases(Clinicalguidelinesontheidentification,evaluation,andtreatment ofoverweightandobesityinadults).Theevidencereport.Bethesda,Md.: NationalHeart,Lung,andBloodInstituteincooperationwiththeNational InstituteofDiabetesandDigestiveandKidneyDiseases;1998.
[38]NelsonHE,WillisonJR.TherevisedNationalAdultReadingTest-TestManual. Windsor:NFER-Nelson;1991.
[39]Nilsson LG, Nilsson E. Overweight and cognition. Scand J Psychol 2009;50(6):660–7.
[40]ReitanR.ValidityoftheTrailMakingTestasanindicatoroforganicbrain damage.PerceptMotSkills1958;8:271–6.
[41]RobinsonLJ,ThompsonJM,GallagherP,GoswamiU,YoungAH,FerrierIN,etal. Ameta-analysisofcognitivedeficitsineuthymicpatientswithbipolar disor-der.JAffectDisord2006;93(1–3):105–15.
[42]SheehanDV,LecrubierY,SheehanKH,AmorimP,JanavsJ,WeillerE,etal.The Mini-InternationalNeuropsychiatricInterview(MINI):thedevelopmentand validationofastructureddiagnosticpsychiatricinterviewforDSM-IVand ICD-10.JClinPsychiatry1998;59(Suppl.20):22–33.
[43]ThompsonJM,GallagherP,HughesJH,WatsonS,GrayJM,FerrierIN,etal. Neurocognitiveimpairmentineuthymicpatientswithbipolaraffective dis-order.BrJPsychiatry2005;186:32–40.
[44]VolkowND,WangGJ,TelangF,FowlerJS,GoldsteinRZ,Alia-KleinN,etal. InverseassociationbetweenBMIandprefrontalmetabolicactivityinhealthy adults.Obesity(SilverSpring)2009;17(1):60–5.
[45]WaltherK,BirdsillAC,GliskyEL,RyanL.Structuralbraindifferencesand cognitivefunctioningrelatedtobodymassindexinolderfemales.HumBrain Mapp2010;31(7):1052–64.
[46]WechslerD.WechslerAdultIntelligenceScale:WAIS-III.Manual.SanAntonio, TX:ThePsychologicalCorporation;1997.
[47]WingoAP,HarveyPD,BaldessariniRJ.Neurocognitiveimpairmentinbipolar disorderpatients:functionalimplications.BipolarDisord2009;11(2):113–25. [48]YathamLN,KennedySH,O’DonovanC,ParikhS,MacQueenG,McIntyreR,etal. CanadianNetworkforMoodandAnxietyTreatments(CANMAT)guidelinesfor themanagementofpatientswithbipolardisorder:consensusand controver-sies.BipolarDisord2005;7(Suppl.3):5–69.
[49]YoungRC,BiggsJT,ZieglerVE,MeyerDA.Aratingscaleformania:reliability, validityandsensitivity.BrJPsychiatry1978;133:429–35.
[50]ZacharyRA.ShipleyInstituteofLivingScale:revisedmanual.LosAngeles: WesternPsychologicalServices;2000.
C.Y.Yimetal./EuropeanPsychiatry27(2012)223–228 228
