Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. - PDF document

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited. EffectofMelatoninonSleep,Behavior,and CognitioninADHDandChronic Sleep-OnsetInsomnia KRISTIAANB.VAN DER HEIJDEN,P H .D.,MARCELG.SMITS,M.D.,P H .D., EUSJ.W.VANSOMEREN,P H .D.,K.RICHARDRIDDERINKHOF,P H .D., AND W.BOUDEWIJNGUNNING,M.D.,P H .D. ABSTRACT Objective: Toinvestigatetheeffectofmelatonintreatmentonsleep,behavior,cognition,andqualityoflifeinchildrenwith Method: Atotalof105medication-free children,ages6to12years,withrigorouslydiagnosedADHDandchronicsleeponsetinsomniaparticipatedina randomized,double-blind,placebo-controlledtrialusing3or6mgmelatonin(dependingonbodyweight),orplacebofor 4weeks.Primaryoutcomeparameterswereactigraphy-derivedsleeponset,totaltimeasleep,andsalivarydimlight melatoninonset. Results: Sleeponsetadvancedby26.9 T 47.8minuteswithmelatoninanddelayedby10.5 T 37.4 minuteswithplacebo( p .0001).Therewasanadvanceindimlightmelatoninonsetof44.4 67.9minutesinmelatonin andadelayof12.8 T 60.0minutesinplacebo( p .0001).Totaltimeasleepincreasedwithmelatonin(19.8 T 61.9minutes) ascomparedtoplacebo( j 13.6 T 50.6minutes; p =.01).Therewasnosignificanteffectonbehavior,cognition,andquality oflife,andsignificantadverseeventsdidnotoccur. Conclusion: Melatoninadvancedcircadianrhythmsofsleep-wake andendogenousmelatoninandenhancedtotaltimeasleepinchildrenwithADHDandchronicsleeponsetinsomnia; J.Am.Acad.ChildAdolesc. Psychiatry ,2007;46(2):233 Y 241. KeyWords: melatonin,insomnia,attention-deficit/hyperactivitydisorder. Attention-deficit/hyperactivitydisorder(ADHD)is characterizedbyinattention,hyperactivity/impulsivity, orboth,andisoneofthemostcommonpsychiatric disordersofchildhood(AmericanPsychiatricAssocia- tion,1994).Approximatelyonethirdofmedication-free insomnia(SOI)(Corkumetal.,1999;Stein,1999).This persistentdisabilitytofallasleepatthedesiredtimein theeveningmayexacerbatedaytimemood,behavioral, and/orcognitiveproblems(Sadehetal.,2003). Thesafetyandefficacyofmelatonintreatmentfor SOIinchildrenwithoutADHDhavebeendocumented inseveralstudies(Smitsetal.,2001;2003;Vander Heijdenetal.,2005a).Melatoninefficacyhasnotbeen AcceptedAugust7,2006. Drs.VanderHeijdenandGunningarewi terKempenhaeghe, Heeze;Dr.VanderHeijdenisalsowith theMaastrichtInstituteofBrainand Behavior,Maastricht;Dr.SmitsiswiththeCenterforSleep-WakeDisordersand Chronobiology,HospitalGelderseVallei,Ede;Dr.VanSomereniswithThe NetherlandsInstituteforNeuroscienceandtheVUUniversityMedicalCenter, DepartmentsofNeurology,Neurophysiol ogyandMedicalPsychology,Amsterdam; BrainandBehavior(Acacia),theDepartmen tofPsychology,UniversityofAmsterdam, andtheDepartmentofPsychology,LeidenUniversity,Leiden,TheNetherlands. ThisstudywassupportedbytheMaartenKapelleFoundationand FoundationDeDrieLichten.TheauthorsthankPharmaNordformaking availablethetrialmedication,pharmacistIngevanGeijlswijkforthe randomizationandkeepingofthecodes,thelaboratoryofGelderseVallei Hospitalforthesalivarymelatoninassays,andProf.R.BuijsandProf.G. Kerkhofforhelpfulsuggestionsduringthedesignofthisstudy. ArticlePlus(onlineonly)materialsforthisarticleappearontheJournal’sWeb site:www.jaacap.com. V URL:http://www.controlled-trials. com/isrctn.Uniqueidentifier:InternationalStandardRandomisedControlled TrialNumberRegister(ISRCTNR)47283236. CorrespondencetoDr.MarcelG.Smits,HospitalGelderseVallei,P.O.Box 9025,6710HNEde,TheNetherlands;e-mail:SmitsM@zgv.nl. 0890-8567/07/4602-0233
2007bytheAmericanAcademyofChild andAdolescentPsychiatry. DOI:10.1097/01.chi.0000246055.76167.0d 233 J.AM.ACAD.CHILDADOLESC.PSYCHIATRY,46:2,FEBRUARY2007 Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited. studiedinmedication-freechildrenwithADHDand SOI;however,thispatientgroupisofspecialinterest. First,medication-freechildrenwithADHDandSOI haveadelayedeveningincreaseinendogenousmelato- ninlevels(VanderHeijdenetal.,2005b)andthisphase delaypredictedastrongersleepphase Y normalizing effectofexogenousmelatonininchildrenwithout ADHD(VanderHeijdenetal.,2005a).Second, becausetreatmentofsleep-relateddisordersotherthan insomniaimproveddaytimefunctioninchildrenwith ADHD(Alietal.,1996;Waltersetal.,2000),the treatmentofinsomniamaybeexpectedtodosoaswell andthusmayhaveimportantconsequencesforADHD treatmentstrategiesinhealthcaresettings. Weinvestigatedwhethermelatonintreatmentcould improveobjectiveandsubjectivemeasuresofsleep, behavior,cognitiveperformance,andqualityoflife inmedication-freechildrendiagnosedwithADHDand SOI. METHOD Subjects Atotalof176childrenwithpossibleADHDwerereferredfor participationinthistrialtooutpatientclinicsforsleep-wakedis- ordersoftheGelderseValleiGeneralHospitalandKempenhaeghe bysevenDutchcommunitymentalhealthinstitutionsandthree pediatrichospitaldepartments.Twenty-eightchildrenwere recruitedthroughadvertisementsinmagazines,newspapers,orvia theDutchADHDpatientsupportcenter.Ofthese204children, 107enrolledinthepresentstudyafterdiagnosticevaluation (Fig.1).Inclusioncriteriawere6to12yearsold,diagnosisof ADHDandSOI,andwritteninformedconsentobtainedfrom parents.ExclusioncriteriaweretotalIQpervasivedevelop- mentaldisorder,chronicpain,knowndisturbedhepaticorrenal function,epilepsy,earlieruseofmelatonin,anduseofstimulants, neuroleptics,benzodiazepines,clonidine,antidepressants,hyp- notics,or " -blockerswithin4weeksbeforeenrollment. ClinicalAssessment All204childrenwereassessedbyapsychologist(K.B.H.)anda board-certifiedchildandadolescentpsychiatrist(W.B.G.).ADHD wasdiagnosedinaccordancewithguidelinesoftheAmerican AcademyofPediatrics(2000)andAmericanAcademyofChildand AdolescentPsychiatry(1997)andincludedclinicalhistory, DiagnosticInterviewScheduleforChildren-Parentform(Shaffer etal.,1996),ChildBehaviorChecklist(CBCL;Achenbach,1991a), andTeacher ` sReportForm(TRF;Achenbach,1991b).Subtypesof ADHDweredeterminedaccordingto DSM-IV .AshortenedIQ test(WISC-RDutchversion:BlockDesign,Vocabulary[DeBruin etal.,1986])wasadministeredwhenIQhadnotbeenassessed previouslyandschoolperformancehadbeensubaverageinthepast 3years. SOIwasdefinedas(1)complaintsofsleep-onsetproblems expressedbyparentsand/orchild,(2)occurrenceonatleast4days/ weekforlongerthan1year,(3)averagesleeponsetlaterthan 8:30 PM forchildrenatage6yearsandforolderchildren15minutes laterperyear,and(4)averagesleeplatencyexceeding30minutes (VanderHeijdenetal.,2005a).Thediagnosticproceduresincluded clinicalhistory,1-week24-houractigraphymeasurement,Dutch SleepDisordersQuestionnaire(Sweereetal.,1998),andChildren ` s SleepHygieneScale(Harshetal.,2002). StudyDesignandConduct A4-weekrandomized,double-blind,placebo-controlledstudy, immediatelyfollowinga1-weekbaselineperiod,wasconducted betweenNovember2001andJune2005.Theprotocolwas approvedbytheinstitutionalreviewboardateachcenter,asa multicentertrialbytheCentralCommitteeonResearchInvolving HumanSubjects,andregisteredintheInternationalStandard RandomizedControlledTrial NumberRegister(ISRCTN- 47283236).Thetrialwasperformedaccordingtothe1997 EuropeanGuidelinesforGoodClinicalResearchPracticein childrenandfollowedthe1983revisedprovisionsofthe1975 DeclarationofHelsinki. Participantswererandomlyassigned(1:1ratio)toreceive melatonin(3mgwhenbodyweightkg[ n =44];6mgwhen bodyweight�40kg[ n =9])infast-releasetablets(PharmaNord, Denmark)oridentical-appearingplacebotabletsat7:00 PM .These pharmacologicaldoseshadpreviouslybeenprovedeffectiveandsafe inchildren(Smitsetal.,2001,2003).Treatmentadherencewas assessedafter3weeksoftreatmentbymedicationcounts.Allofthe measurementstookplaceatbaselineandduringthefourth treatmentweek.Parentswerecalledweeklytomonitorprogress anddiscusspossibleproblems.Childrenwereallowedtogotobed whenevertheyfelttiredratherthanbeingtiedtoascheduled bedtime.Participantswerenotallowedtostartorchange therapeuticinterventionsduringthestudy.Noassessmentstook placeduringorwithin3daysfollowingholidaysorclocktime transitions.Alloftheparticipantswereofferedmelatonintreatment aftertheendofthestudy. OutcomeMeasures Sleep. Sleepwasestimatedusingactigraphy(Actiwatch,Cambridge NeurotechnologyLtd.,Cambridge,UK)andsleeplogsonseven consecutivedays,atbaseline,andduringthefourthtreatmentweek. Actigraphs,wornonthenondominantwrist,recordedtheamountof movementat1-minuteepochs,24hours/day.Actigraphydatawere convertedintosleepparametersbythevalidatedautomatic Actiwatchscoringalgorithm(Kushidaetal.,2001),usingsleep log Y derivedlightsoutandrisetime,andbysubsequentmanual verificationbasedonsleeplogdata:sleeponset,sleeplatency (timefromlightsoutuntilsleeponset),wakeuptime;total timeasleep,sleepefficiency,movingtime(percentagetimespent movingduringassumedsleepperiod).Furthermore,nonpara- metricvariableswerederivedfromactigraphydataasdescribed previously(VanSomerenetal.,1999):interdailystability (degreeofresemblancebetweentheactivitypatternson individualdays),intradailyvariability(fragmentationofperiods ofsleepandactivity),L5(averageactivityduringtheleast-active 5hours).(Detailedinformationonoutcomeparametersis availableviatheArticlePlusfeatureatthe Journal’s Website VANDERHEIJDENETAL. 234 J.AM.ACAD.CHILDADOLESC.PSYCHIATRY,46:2,FEBRUARY2007 Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited. www.jaacap.com ).Primarysleepoutcomeparametersweresleep onset,totaltimeasleep,andsleeplogitemdifficultyfallingasleep (averagedover7days,onascaleof1[notdifficult]to5[very difficult]). DimLightMelatoninOnset(DMLO). DLMOistheclocktimeat whichtheendogenousmelatoninlevelstartstoriseinthesubjective eveningandisconsideredthemostreliablephasemarkerofthe biologicalclockrhythm(Klermanetal.,2002).DLMOwasassessed atbaselineandonthefirsteveningofthefourthtreatmentweek. Hourlysamplesofsalivafrom6:00to10:00 PM (6 Y 7yearsold),or 7:00to11:00 PM (8to12yearsold)wereobtainedbychewingona cottonplugfor1minute(Salivetten,SarstedtNu ¨ mbrecht, Germany).Nomedicationwastakenontheeveningofthesalivary samplecollectionbecausemedicationintakemayalterthe endogenouslevelsofmelato nin.Topreventsuppressionof melatoninsecretionbybrightlight,curtainsneededtobeclosed, andonlyonedimlightwasallowedduringtheentiremeasurement period.DLMOwasdefinedasthelinearlyinterpolatedtimeat whichthemelatoninconcentrationfirstreached4pg/mL. Melatoninconcentrationsweremeasuredasdescribedpreviously (Nagtegaaletal.,1998). ProblemBehavior,CognitivePerformance, andQualityofLife Theprimaryprespecifiedoutcomeparameterwastheaveraged grade(1=verysevere;10=none)onthreeindividuallydefined (spontaneously,notfromachecklist),mostserious,andcommon coreproblemsofthechild,givenbyparentsaswellasteachers. EmotionalandbehavioralproblemswereassessedwiththeCBCL andTRF,whichconsistedof120items(responsescale:0=nottrue; 1=somewhatorsometimestrue;2=verytrueoroftentrue),and yieldingatotalscoreandeightsyndromescalescores(highscores indicatingmoreproblems). Interferencecontrolistheabilitytoignoreinformationlinkedto aninappropriateresponsetendencyandwasshowntobeimpaired inADHD(Cornoldietal.,2001).TheEriksentaskisa Fig.1 Participantflowdiagram.SOI=sleep-onsetinsomnia. MELATONINTREATMENTINADHD 235 J.AM.ACAD.CHILDADOLESC.PSYCHIATRY,46:2,FEBRUARY2007 Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited. computerizedchoicereactiontime(RT)taskofinterferencecontrol. Subjectshadtorespondonthedirectionofatargetarrow,while, occasionally,aconflictoccurredbetweenthisdesignatedresponse andacompetingresponsetendencyelicitedbyincongruentflanking arrows,whichtheyhadtoignore(6blocksof60stimuli,ofwhich1 wasapracticeblock).Comparedwiththecongruentandneutral flankers,theconflictinginformationproducedbytheincongruent flankershasbeenreportedtoleadtoincreasederrorincidence(EI $ , percentage)anddelayedRTs(RT $ ,milliseconds[Ridderinkhof etal.,2005]). SustainedattentionwasshownimpairedinADHDwith improvementsafterstimulanttreatment(Riccioetal.,2001).We administeredtheSustainedAttentionDotsTask(Amsterdam NeuropsychologicalTasks[DeSonneville,1999]),acomputerized visualtaskinwhichsubjectswerepresentedacontinuousand consecutiveseriesof300target(33%)andnontarget(67%) asymmetricdotconfigurationsinapseudo-randomfashion. Subjectspusheda B yes ^ buttonwheneveratargetappearedanda B no ^ buttonafteranontargetsignal,afterwhichthenextstimulus waspresentedimmediately(i.e.,self-paced).Thetaskparameters wereinaccuracy(%)andtaskcompletiontime(seconds). TheTNO-AZLQuestionnaireforChildren ` sHealth-Related QualityofLife,ParentForm(TACQOL-P;Vogelsetal.,2000) consistedof63itemsonwhetherspecificproblemsoccurredor feelinghadbeenpresentinrecentweeks,witha3-pointLikert responsescale,yieldingatotalsumscore(maximum,224:highest qualityoflife)aswellasscoresonsevensubdomains. AdverseEvents Parentsreportedadverseeventsinanunstructuredinterview (K.B.H.)3weeksafterthestartofstudytreatment.A2-yearfollow- upwasconductedusingaself-constructed,structuredquestionnaire withitemsonvarioustreatmentaspectsandadverseeffects. DataAnalysis Atotalof107patientswasselectedforenrollment.(samplesize calculationisavailableviatheArticlePlusfeatureatthe Journal’s Website www.jaacap.com ).Randomizationwasperformedbya hospitalpharmacistnotconnectedtothestudyinblocksoffourto keepthenumberofpatientsineachtreatmentgroupclosely balancedatalltimes.Thefollowingstratificationcriteriawereused: (1)presenceofpsychiatriccomo rbidity(disruptivebehavior disorder[ n =59];anxietydisorder[ n =16];depressivedisorder [ n =1]),(2)agecategory(6 Y 9years[ n =66];10 Y 12years[ n =39]), and(3)bodyweight(kg[ n =88];=40kg[ n =17]). Investigatorsandparticipantswereunawareoftreatmentallocation. Thecodewasbrokenafterallofthechildrencompletedtreatment anddatawererecorded(October2005). Comparisonsofdemographicandclinicalcharacteristicsbetween treatmentgroupswereconductedusingindependentsamples t test forcontinuousvariableswithanormaldistribution,Mann-Whitney U testwhendistributionwasnotnormal,andPearson x 2 testfor discretecategoricalvariables.Between-groupdifferencesinmean daylength(hours)andmeanrateofchangeindaylength(hoursper week)wereanalyzedtocontrolforpossibleconfounding. Analysesofbetween-groupdifferencesinpre-toposttreatment changeswereconductedusinggenerallinearmodel(GLM) repeated-measuresanalysisofvarianceforcontinuousvariable, andwithPearson x 2 testsforcategoricalvariables.Therelationship ofpre-toposttreatmentchangeswithvariablesofinterestwas analyzedwithlinearregressionanalysis.Between-groupdifferences inadverseeventswereanalyzedwiththeFisherexacttest. AnalyseswereconductedusingSPSS,12.0.1(SPSS,Inc., Chicago,IL)onanintention-to-treatbasis(significance p =.05, two-sided).TocompensatefortheincreasedprobabilityofatypeI error,Bonferronicorrectionof p valueswasappliedwhenmultiple testswereconductedonquestionnairesubscales(CBCL,TRF, TACQOL),calculatedbymultiplyingtheresulting p valuesbythe numberofoutcomesbeingtested(correctionsareindicated). RESULTS BaselineDemographicandClinicalCharacteristics Of107patients,105receivedmelatonin( n =53)or placebo( n =52).Twowerewithdrawnbecauseshortly afterassignmenttheystartedanothertreatmentwithout permission(Fig.1).Atbaseline,therewereno significantbetween-groupdifferencesindemographic variables,clinicalcharacteristics,meandaylength,or meanrateofchangeindaylength(Table1). TABLE1 DemographicandClinicalCharacteristics Melatonin ( n =53) Placebo ( n =52) p Value Age,mean T SD(y)9.1 T 2.39.3 T 1.8.64 a Bodyweight(kg)31.8 T 10.332.7 T 9.7.67 a Male,no.(%)35(66.0)43(82.7).051 b ADHDsubtype,no.(%) ADHD-C41(77.4)36(69.2).34 b ADHD-I9(17.0)13(25.0).31 b ADHD-HI2(3.8)2(3.8).98 b Psychiatriccomorbidity(%) Disruptive behavioraldisorder 31(58.5)28(53.8).71 b Anxietydisorder7(13.2)9(17.3).53 b Depressivedisorder1(1.9)0(0).32 b Insomniascore(SDQ)2.6 T 0.52.6 T 0.6.69 c PLMS/RLSscore(SDQ)1.6 T 0.71.6 T 0.6.84 c OSASscore(SDQ)1.2 T 0.31.2 T 0.3.81 c Sleephygienescore(CSHS)55.4 T 10.257.2 T 10.2.40 c Note: ADHD=attention-deficit/hyperactivitydisorder;ADHD- C=attention-deficit/hyperactivitydisordercombinedsubtype; ADHD-I=attention-deficit/hyper activitydisorderinattentive subtype;ADHD-HI=attention-de ficit/hyperactivitydisorder hyperactive/impulsivesubtype;SDQ=DutchSleepDisorders Questionnaire(scale1=neverto5=always);PLMS/RLS=periodic limbmovementdisorder/restlesslegssyndrome;OSAS=obstructive sleepapnea;CSHS=Children ` sSleepHygieneScale(scale25 Y 150, highervaluesindicateworsesleephygiene). a Independentsamples t test. b Pearson x 2 test. c Mann-Whitney U test. VANDERHEIJDENETAL. 236 J.AM.ACAD.CHILDADOLESC.PSYCHIATRY,46:2,FEBRUARY2007 Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited. EfficacyMeasures Pre-aswellasposttreatmentdatawereavailablefrom amean( T SD)of69.6% T 17.4%(range,32.4% Y 86.7%)oftheparticipants,varyingwithoutcome measure.(RawdataareavailableviatheArticlePlus featureatthe Journal’s Website www.jaacap.com ). Missingdatamainlyresultedfromtechnicalproblems (actigraphy),insufficientvolumeofcollectedsaliva (DLMO),technicalproblemsorrefusalofchildrento performthetask(cognitiveperformance),partial completionofquestionnaires,lossbyparticipants, andnonadherencetoinstructions. Sleep Meanactigraphicestimateofsleeponsetadvancedby 26.9 T 47.8minuteswithmelatonin,whereastherewas adelayof10.5 T 37.4minuteswithplacebo( p .0001; Table2).Aftermelatonin,20(48.8%)of41childrenfor whomactigraphydatawereavailableshowedanadvance ofsleeponset�30minutes,whereasthiswas5of39 (12.8%)afterplacebo( x 2 =12.0; p =.001).Therewasan increaseinmeantotaltimeasleepof19.8 T 61.9minutes withmelatoninandadecreaseof13.6 T 50.6minuteswith placebo( p =.01).Ascomparedwithplacebo,the melatoningroupshowedasignificantdecreaseinsleep latency( p =.001),increaseinsleepefficiency( p =.01),and decreaseofnocturnalrestlessness(L5; p =.03).Between- groupdifferencesinchangesofotheractigraphicsleep measureswerenotsignificant.Themeanscoreonsleep logitemdifficultyfallingasleepdecreasedby1.2 T 1.3 points(35.3%ofbaseline)withmelatoninandby0.1 T 0.8points(4.3%ofbaseline)withplacebo( p .0001). TABLE2 OutcomeVariables Variable MelatoninPlacebo p Value a BaselineTreatmentChangeBaselineTreatmentChange DLMO(hr:min)20:37 T 0:5619:53 T 1:06 j 0:44 T 1:0720:32 T 0:5420:45 T 1:06+0:13 T 0:59 Sleep Sleeponset(hr:min)21:40 T 0:5921:13 T 0:58 j 0:27 T 0:4821:38 T 0:4721:48 T 0:48+0:10 T 0:37 Sleeplatency(min)53.0 T 22.031.7 T 30.7 j 21.3 T 33.047.3 T 23.250.4 T 30.4+3.0 T 31.7.001* Totaltimeasleep(min)518.9 T 48.3538.7 T 55.0+19.8 T 61.9533.8 T 47.8520.2 T 47.5 j 13.6 T 50.6.01* Sleepefficiency(%)80.1 T 5.582.7 T 7.5+2.6 T 8.982.4 T 5.780.3 T 5.9 j 2.1 T 7.1.011* Difficultyfallingasleep b 3.4 T 0.92.2 T 0.9 j 1.2 T 1.33.2 T 0.73.1 T 1.0 j 0.1 T 0.8 Problembehavior Coreproblems_ parents 4.0 T 1.34.8 T 1.3+0.7 T 0.94.3 T 1.34.5 T 1.2+0.2 T 0.8.002* Coreproblems_ teacher 4.6 T 1.05.3 T 1.1+0.7 T 0.94.4 T 1.24.9 T 1.4+0.5 T 1.0.41 CBCL63.0 T 16.055.1 T 18.4 j 8.0 T 8.861.5 T 21.845.3 T 25.7 j 16.2 T 18.2.083 TRF46.1 T 20.742.1 T 19.1 j 4.0 T 12.552.2 T 26.148.1 T 25.0 j 4.0 T 16.4.29 Qualityoflife TACQOL-P_ total 170.4 T 19.9179.1 T 21.8+8.7 T 13.0168.8 T 22.5176.9 T 22.5+8.1 T 9.1.82 Cognition Interferencecontrol(EFT) $ Reactiontime c (ms)50.1 T 37.133.1 T 33.0 j 17.0 T 43.241.8 T 27.832.3 T 25.4 j 9.5 T 34.4.37 $ Errorincidence d (%)3.4 T 4.03.8 T 3.5+0.4 T 4.53.8 T 3.73.2 T 3.5 j 0.6 T 3.6.28 Sustainedattention(SADT) Inaccuracy(%) e 12.6 T 7.612.2 T 7.2 j 0.3 T 4.512.0 T 6.712.34 T 7.3+0.4 T 5.6.52 Taskcompletiontime(s)469.0 T 153.0410.5 T 142.6 j 58.5 T 77.8489.4 T 187.1428.2 T 171.9 j 61.2 T 85.5.58 Note: DLMO=salivarydimlightmelatoninonset;CBCL=ChildBehaviorChecklist,totalscore;TRF=Teacher ` sReportForm,total score;TACQOL-P=TNO-AZLQuestionnaireforChildren ` sHealth-RelatedQualityofLife,ParentForm;EFT=EriksenFlankerTask; SADT=SustainedAttentionDotsTask(AmsterdamNeuropsychologicalTasks). a Generallinearmodelrepeated-measurestestongroupdifferencesinpre-toposttreatmentchanges. b Meandifficultyfallingasleepasreportedbyparents,averagedover7daysonascaleof1(notdifficult)to5(verydifficult). c Differenceinreactiontimebetweenincongruentandcongruenttrials. d Differenceinerrorincidencebetweenincongruentandcongruenttrials. e Percentageofmissesplusfalsealarmsrelativetothetotalnumberoftrials. *Statisticallysignificantdifference( p =.05,two-sided). MELATONINTREATMENTINADHD 237 J.AM.ACAD.CHILDADOLESC.PSYCHIATRY,46:2,FEBRUARY2007 Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited. DMLO Melatonin-treatedchildrenshowedanadvancein DLMOof44.4 T 67.9minutescomparedwitha delayof12.8 T 60.0minutesinchildrenreceiving placebo( p .0001).Inmelatonin,pre-toposttreat- mentchangesinsleeponsetshowedasignificantlinear relationshipwithpretreatmentvaluesofDLMO( R =0.42; p =.008),indicatingthatmoredelayedDLMOvaluesat baselineassociatedwithstrongeradvancesofsleeponsetafter melatonintreatment.Thisrel ationshipwasnotsignificant inplacebo( R =0.078; p =0.645).Inmelatonin,pre-to posttreatmentchangesinDL MOwerenotsignificantly relatedwithpre-toposttreatmentchangesinsleeponset ( R =0.30; p =.124).Between-groupdifferencesinchanges ofDLMOwerenotsignificantlyrelatedtopresenceof comorbidpsychiatricdisorders. ProblemBehavior,CognitivePerformance, andQualityOfLife CoreProblems. Themeanscoreatbaselinewas4.0 T 1.3 inmelatoninand4.3 T 1.3inplacebo(maximum,10:no problems).Themostfrequentlyreportedproblemswere easytoanger(36.0%),sleep-onsetproblems(31.8%), andattentionproblems(23.3%).Meanscoreimproved by0.7 T 0.9(18.4%ofbaseline)withmelatoninbyand 0.2 T 0.8(3.7%ofbaseline)withplacebo( p =.002).This groupdifferenceloststatisticalsignificanceafterremov- ingcoresymptomsrelatedtosleep.Between-group differencesinchangesofteacher-reportedcoreproblems werenotsignificant. BehavioralandEmotionalSymptoms. Between-group differencesinpre-toposttreatmentchangesofmeantotal CBCL( p =.083)andTRF( p =.294)scoreswerenot significant.Improvementsontheaggressivebehavior subscaleofCBCLweresignificantlysmallerwith melatonin( j 0.6 T 3.4)comparedwithplacebo( j 6.2 T 6.9;Bonferronicorrected p =.024).Changesintotal CBCLandTRFscoreswerenotrelatedtochangesin actigraphy-derivedsleeponset,totaltimeasleep,orsleep log Y deriveddifficultyfallingasleep. InterferenceControl. Atbaseline,meanRTonincon- gruenttrialswassignificantlyslower(696.8 T 146.4 seconds)thanoncongruenttrials(650.8 T 140.1seconds; p .0001),andEIonincongruenttrials(7.8% T 6.4%) wassignificantlyhigherthanoncongruenttrials(4.2% T 4.2%; p .0001).Thesedataindicateasignificanteffect ofcongruenceoninterfer encecontrol.Melatonin significantlyreducedRT $ ofcorrectresponseswith 17.0 T 43.2milliseconds,indicatinganimprovementof interferencecontrol;however,thedifferencewith placebo(9.5 T 34.4milliseconds)wasnotsignificant ( p =.37).Betweengroup-differencesinchangesofEI $ werealsonotsignificant( p =.28). SustainedAttention. Neithermeantaskcompletion timenorinaccuracychangedsignificantlywithmelatonin ascomparedtoplacebo( p =.58and p =.52,respectively). QualityofLife. MeantotalTACQOL-Pscoreat baselinewas170.4 T 19.9inmelatoninand168.8 T 22. 5inplacebo(maximum,224).Neitherbetween-group differencesinchangesoftotalTACQOL-Pscorenor Bonferroni-correctedchangesonanyoftheseven subscalesweresignificant.Therewasnosignificant relationshipofchangesintotalTACQOL-Pscorewith changesinactigraphy-derivedsleeponset,totaltime asleep,orsleeplog Y deriveddifficultyfallingasleep. AdverseEvents Thenumberofadverseeventsdidnotdiffer significantlybetweenmelatoninandplacebo(Table3) norbetweenthe3-mg(8/44)and6-mg(2/9)treated groups( p =1.00).Fivepatientshadoneadverseevent, fourpatientshadtwo,andonehadthreeadverseevents. Therewerenodiscontinuationsorwithdrawalscaused byadverseevents,andnoneoftheadverseevents requiredtreatment. TABLE3 AdverseEvents Adverseevent Melatonin( n =53), No.(%) a Placebo( n =52), No.(%) a p Value b Headache3(5.7)0(0).24 Hyperactivity3(5.7)0(0).24 Dizziness2(3.8)0(0).50 Abdominalpain2(3.8)0(0).50 Nosebleeding1(1.9)0(0)1.00 Itchinglumpson theskin 1(1.9)0(0)1.00 Painfullumpson theskin 1(1.9)0(0)1.00 Diarrhea1(1.9)0(0)1.00 Decreaseof mood 1(1.9)0(0)1.00 Maintenance insomnia 1(1.9)0(0)1.00 a Patientscouldreportmorethanoneevent. b Fisherexacttest. VANDERHEIJDENETAL. 238 J.AM.ACAD.CHILDADOLESC.PSYCHIATRY,46:2,FEBRUARY2007 Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited. Follow-upat2yearsafterparticipationyielded24of26 completedquestionnaires(2familieswereuntraceable): 19of24stillusedmelatonin(4.4 T 2.0mgat7:42 PM T 50minutes),1useditoccasionally,and4stoppedafter 17.23 T 3.3months(reasons:remission[ n =3], dizzinessanddrowsiness[ n =1]).Sevenof24parents reportedoneormoreofthefollowingadverseevents: bedwetting( n =2),abnormalfeces( n =2),drowsiness ( n =2),dizziness( n =1),sleepmaintenanceproblems ( n =1),skinpigmentchanges( n =1),anddecreased mood( n =1). DISCUSSION Thisisthefirstrandomized,placebo-controlledtrial toinvestigatebenefitsandharmsofmelatoninin medication-freechildrenwithrigorouslydiagnosed ADHDandSOI.Melatoninimprovedobjectivesleep onsetandsleepduration,reducedsubjectivedifficulty fallingasleep,andinducedadvancesofsleeponsetof �30minutesinabouthalfofthemelatonin-treated children.However,thefindingsdidnotsupportour initialhypothesisthatmelatonintreatmentimproved problembehavior,cognitiveperformance,orqualityof life.Althoughparentsreportedanimprovementin severityofindividuallydefinedcoreproblems,thiswas mainlytheresultofanameliorationofproblemsrelated tosleep.Melatoninusewasnotassociatedwith significantadverseevents,whichcorrespondswith previousrandomizedclinicaltrialsininsomniac children(DodgeandWilson,2001;Janetal.,1994; 2000;McArthurandBudden,1998;O ` Callaghan etal.,1999;Smitsetal.,2001;2003). Melatoninadvancedsleeponset(27minutes)and DLMO(44minutes)tonormalvaluesfoundpreviously inchildrenwithADHDwithoutinsomnia(Vander Heijdenetal.,2005b)orhealthychildren(Vander Heijdenetal.,2005a).Thepresentresultscorroborate ourpreviousfindingthatdelayedvaluesofpretreatment DLMOareassociatedwithmorerobustadvancementsof sleep-wakerhythmaftermelatonintreatment(Vander Heijdenetal.,2005a).Thesefindingssuggestthat melatoninsynchronizedcircadianrhythmsregulatedby thebiologicalclock.However,becausewefoundthat melatonin-inducedchangesinDLMOdidnotrelate significantlytochangesinsleeponset,itislikelythatalso directsoporificeffectswereinvolvedintheimproved sleepinitiation(Sacketal.,1997;VanSomeren,2000). Inspiteofimprovementsinsleep,melatonin demonstratednoeffectonbehavior,cognitive performance,andqualityoflife.Weexpectedsuch improvementsbecausesleepproblemsandsleep deprivationinchildrenwereassociatedwithbeha- vioraldisturbances(Dahl,1996;Sadehetal.,2003). Furthermore,treatmentofchildhoodinsomnia showedimprovementsinhealthstatus(Smitsetal., 2003),althoughthiswasnotfoundforsustained attention(Smitsetal.,2001).Possibleexplanations arethatthepresentimprovementsinsleepwerenot largeenoughtoinduceimprovementsinbehavior, cognitiveperformance,andqualityoflife;theymay haverequiredlongertreatmentduration;ortheywere possiblymaskedbylargecognitivedeficitsfoundin childrenwithADHD. Thenegativeplaceboeffectonobjectivesleep measureswasconsistentwithpreviousresultsin insomniacchildren(Smitsetal.,2001;Vander Heijdenetal.,2005a),butnotwithpositiveplacebo effectsusuallyfoundinadultinsomniacs(Perlisetal., 2005).Thisisaninterestingphenomenonforwhichwe donothaveaclearexplanation. Thefindingthatnoadverseeventswerereportedin theplacebo-treatedgroupisunusualandmayrelateto thenegativeeffectsofplaceboonsleep;namely,parents whoexperiencedaworseningofthechild ` ssleep problemassumedthattheirchildreceivedplacebo, whichmayhavecausedabiastowardnotperceivingor reportingadverseevents. Althoughweusedpharmacologicaldosesof3or 6mg,previousstudiesinchildrenusedpharmacological dosesof2to12mg,whichwerefoundtobesafeand effective(Janetal.,1994,2000;McArthurandBudden, 1998;Smitsetal.,2001,2003).Physiologicaldosesof melatonin(0.1,0.3,and1.0mg)showedsleep- promotingeffectsinnormalaswellasinsomniac adults(AlmeidaMontesetal.,2003;Zhdanovaetal., 1995),buthavenotyetbeenstudiedadequatelyin children. Strengthsofthepresentstudyweretherelativelylarge samplesize,applicationofrigorousdiagnosticmethods andstrictdiagnosticcriteriaforADHDaswellasSOI, andstratifiedrandomizationtocontrolfortheeffectof importantfactors.Furthermore,weusedobjectiveas wellassubjectivemeasuresofsleep,behavior,cognitive performance,andqualityoflifeandconducteda systematic2-yearfollow-upofadverseevents. MELATONINTREATMENTINADHD 239 J.AM.ACAD.CHILDADOLESC.PSYCHIATRY,46:2,FEBRUARY2007 Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited. Limitations Actigraphyisawell-validatedinstrumenttoevaluate sleep-wakerhythmoverlongperiodsinchildren (Littneretal.,2003);however,polysomnography assessmentsarerequiredtoevaluatesleepqualityor sleep-relateddisorderssuchasperiodiclimbmovement disorderandsleep-disorderedbreathing.Nevertheless, inthepresentstudy,reportsofsleepdisordersymptoms wererareandequallydistributedovertreatment groups,whichreducestheriskthatthisfactoraffected theresults.Furthermore,clinicallyrelevanteffectsof melatoninonsleepqualityarenotexpected(Almeida Montesetal.,2003).Thenumberofmissingdataon somemeasuresofbehaviorandqualityoflifewas relativelylarge;however,theriskthatareporterbias distortedthedataissmallbecausetheamountof missingdatawasequalinbothtreatmentgroups. Thepresentfindingsarelikelygeneralizabletoother ADHDpopulationsbecausewerecruitedfromhetero- geneoushealthcaresettings,includedallADHD subtypes,anduseddiagnosticmethodsaccordingto widelyusedpsychiatricc lassificationsystemsand practiceguidelines.Ourfindingsmaynotpertainto stimulant-treatedchildrenwithADHDbecauseevi- dencesuggeststhatstimulanttreatmentcandeteriorate sleep(Ahmannetal.,1993;Stein,1999).Adouble- blind,placebo-controlledtrialinstimulant-treated childrenshowedareductioninsleep-onsetinsomnia of16minuteswithmelatonincomparedtoplacebo (Weissetal.,2006).Anopen-labelstudyinstimulant- treatedchildrenshowedrobustadvancesofsleeponset (median,135minutes),withoutsignificantadverse events(TjonPianGietal.,2003). ChildrenwithADHDshowedagreaternight-to-night variabilityofsleeppatternscomparedwithnormal controls(Gruberetal.,2000).Wefoundaninterdaily stabilityofsleep-wakerhythmof0.65to0.64,which wassimilartothe0.63foundpreviouslyinchildrenwith ADHDwithoutinsomnia(VanderHeijdenetal., 2005b),butwasslightlylowerthaninchildrenwithout ADHDorinsomnia(0.68 T 0.13; n =9;unpublished results).Itremainsunknownwhetherthisfactormay haveaffectedthepresentresults. OurcriteriaofSOIwerebasedonaDutchchild populationbecauseinternationalconsensuscriteriawere notavailable.Thesecriteriaconformedtoorwereeven stricterthanrecentconsensuscriteriaforadultinsomnia (AmericanAcademyofSleepMedicine,2001;Edinger etal.,2004).Itremainsunknownwhetherthecurrent resultsalsopertaintoothercultures. ClinicalImplications Melatonininducedclinicallyrelevantadvancesof sleeponsetandincreasedtotaltimeasleepinchildren withADHDandSOI,withnoapparenteffectson behavior,cognition,andqualityoflifeandwithno significantadverseevents.Nevertheless,werecom- mendthatmelatonintreatmentbeprescribedonly whencomplaintsofinsomniaarepersistentandsevere andimposeaburdenontheindividualchild,if possibleafterameliorationofpossibleunderlying extrinsicfactorsandpreferablyinthosechildren demonstratingadelayedonsetofendogenousmela- toninrhythm.Systematicstudiesonpossiblelong- termeffects,suchasonthegonadotropicsystemand onsetofpuberty(LuboshitzkyandLavie,1999; Reiter,1998)havenotbeenconducted.Furthermore, melatoninhasshownproconvulsantpropertiesin childrenwithepilepsy(Sheldon,1998),although anticonvulsanteffectswerefoundaswell(Peledetal., 2001).Previousstudiesincludingchildrenyoungerthan 6yearsofageandadolescentsdemonstratedefficacyand safetyofmelatonin(DodgeandWilson,2001);however, rigoroustrialsinthesespecificagegroupsstillmustbe conducted. Disclosure:Theauthorshavenofinancialrelationshipstodisclose. REFERENCES AchenbachTM(1991a), ManualfortheChildBehaviorChecklist/4-18and 1991Profile .Burlington:UniversityofVermont,Departmentof Psychiatry AchenbachTM(1991b), ManualfortheTeacher ` sReportFormand1991 Profile .Burlington:UniversityofVermont,DepartmentofPsychiatry AhmannPA,WaltonenSJ,OlsonKA,TheyeFW,VanEremAJ,LaPlantRJ (1993),Placebo-controlledevaluationofRitalinsideeffects. Pediatrics 91:1101 Y 1106 AliNJ,PitsonD,StradlingJR(1996),Sleepdisorderedbreathing:effectsof adenotonsillectomyonbehaviourandpsychologicalfunctioning. EurJ Pediatr 155:56 Y 62 AlmeidaMontesLG,OntiverosUribeMP,CortesSJ,HeinzeMG(2003), Treatmentofprimaryinsomniawithmelatonin:adouble-blind, placebo-controlled,crossoverstudy. JPsychiatryNeurosci 28:191 Y 196 AmericanAcademyofChildandAdolescentPsychiatry(1997),Practice parameterfortheassessmentandtreatmentofchildren,adolescents,and adultswithattention-deficit/hyperactivitydisorder. JAmAcadChild AdolescPsychiatry 36:85S Y 121S AmericanAcademyofPediatrics(2000),Clinicalpracticeguideline: diagnosisandevaluationofthechildwithattention-deficit/hyperactivity disorder. Pediatrics 105:1158 Y 1170 VANDERHEIJDENETAL. 240 J.AM.ACAD.CHILDADOLESC.PSYCHIATRY,46:2,FEBRUARY2007 Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited. AmericanAcademyofSleepMedicine(2001), ICSD Y International ClassificationofSleepDisorders,Revised:DiagnosticandCodingManual . Chicago:AmericanAcademyofSleepMedicine AmericanPsychiatricAssociation(1994), DiagnosticandStatisticalManual ofMentalDisorders,4thedition(DSM-IV) .Washington,DC:American PsychiatricAssociation CorkumP,MoldofskyH,HoggjohnsonS,HumphriesT,TannockR (1999),Sleepproblemsinchildrenwithattention-deficit/hyperactivity disorder:impactofsubtype,comorbidity,andstimulantmedication. JAmAcadChildAdolescPsychiatry 38:1285 Y 1293 CornoldiC,MarzocchiGM,BelottiM,CaroliMG,MeoT,BragaC (2001),Workingmemoryinterferencecontroldeficitinchildren referredbyteachersforADHDsymptoms. NeuropsycholDevCognC ChildNeuropsychol 7:230 Y 240 DahlRE(1996),Theimpactofinadequatesleeponchildren ` sdaytime cognitivefunction. SeminPediatrNeurol 3:44 Y 50 DeBruinE,VanderSteeneG,vanHaasenP(1986), WISC-R:Wechsler IntelligenceScaleforChildren-Revised(Dutchversion),Manual .Lisse,The Netherlands:Swets&Zeitlinger DeSonnevilleLMJ(1999),AmsterdamNeuropsychologicalTasks:A computer-aidedassessmentprogram.In: CognitiveErgonomics,Clinical AssessmentandComputer-AssistedLearning:ComputersinPsychology ,Den BrinkerBPLM,BeekPJ,BrandAN,MaarseSJ,MulderLJM,eds.Lisse, TheNetherlands:Swets&Zeitlinger,pp187 Y 203 DodgeNN,WilsonGA(2001),Melatoninfortreatmentofsleepdisorders inchildrenwithdevelopmentaldisabilities. JChildNeurol 16:581 Y 584 EdingerJD,BonnetMH,BootzinRRetal.(2004),Derivationofresearch diagnosticcriteriaforinsomnia:reportofanAmericanAcademyofSleep MedicineWorkGroup. Sleep 27:1567 Y 1596 GruberR,SadehA,RavivA(2000),Instabilityofsleeppatternsinchildren withattention-deficit/hyperactivitydisorder. JAmAcadChildAdolesc Psychiatry 39:495 Y 501 HarshJR,EasleyA,LeBourgeoisMK(2002),Ameasureofsleephygiene. Sleep 25:A316 JanJE,EspezelH,AppletonRE(1994),Thetreatmentofsleepdisorders withmelatonin. DevMedChildNeurol 36:97 Y 107 JanJE,HamiltonD,SewardN,FastDK,FreemanRD,LaudonM(2000), Clinicaltrialsofcontrolled-releasemelatonininchildrenwithsleep-wake cycledisorder. JPinealRes 29:34 Y 39 KlermanEB,GershengornHB,DuffyJF,KronauerRE(2002),Compar- isonsofthevariabilityofthreemarkersofthehumancircadian pacemaker. JBiolRhythms 17:181 Y 193 KushidaCA,ChangA,GadkaryC,GuilleminaultC,CarrilloO,Dement WC(2001),Comparisonofactigraphic,polysomnographic,and subjectiveassessmentofsleepparametersinsleep-disorderedpatients. SleepMed 2:389 Y 396 LittnerM,KushidaCA,AndersonWMetal.(2003),Practiceparametersfor theroleofactigraphyinthestudyofsleepandcircadianrhythms:an updatefor2002. Sleep 26:337 Y 341 LuboshitzkyR,LavieP(1999),Melatoninandsexhormoneinter- relationships V areview. JPediatrEndocrinolMetab 12:355 Y 362 McArthurAJ,BuddenSS(1998),SleepdysfunctioninRettsyndrome:a trialofexogenousmelatonintreatment. DevMedChildNeurol 40: 186 Y 192 NagtegaalE,PeetersT,SwartW,SmitsM,KerkhofG,vanderMeerG (1998),Correlationbetweenconcentrationsofmelatonininsalivaand seruminpatientswithdelayedsleepphasesyndrome. TherDrugMonit 20:181 Y 183 O ` CallaghanFJ,ClarkeAA,HancockE,HuntA,OsborneJP(1999),Use ofmelatonintotreatsleepdisordersintuberoussclerosis. DevMedChild Neurol 41:123 Y 126 PeledN,ShorerZ,PeledE,PillarG(2001),Melatonineffectonseizuresin childrenwithsevereneurologicdeficitdisorders. Epilepsia 42:1208 Y 1210 PerlisML,McCallWV,JungquistCR,PigeonWR,MattesonSE(2005), Placeboeffectsinprimaryinsomnia. SleepMedRev 9:381 Y 389 ReiterRJ(1998),Melatoninandhumanreproduction. AnnMed 30:103 Y 108 RiccioCA,WaldropJJ,ReynoldsCR,LoweP(2001),Effectsofstimulants onthecontinuousperformancetest(CPT):implicationsforCPTuse andinterpretation. JNeuropsychiatryClinNeurosci 13:326 Y 335 RidderinkhofKR,ScheresA,OosterlaanJ,SergeantJA(2005),Deltaplots inthestudyofindividualdifferences:newtoolsrevealresponse inhibitiondeficitsinAD/HDthatareeliminatedbymethylphenidate treatment. JAbnormPsychol 114:197 Y 215 SackRL,HughesRJ,EdgarDM,LewyAJ(1997),Sleep-promotingeffects ofmelatonin:atwhatdose,inwhom,underwhatconditionsandby whatmechanisms? Sleep 20:908 Y 915 SadehA,GruberR,RavivA(2003),Theeffectsofsleeprestrictionand extensiononschool-agechildren:whatadifferenceanhourmakes. Child Dev 74:444 Y 455 ShafferD,FisherP,DulcanMKetal.(1996),TheNIMHDiagnostic InterviewScheduleforChildrenVersion2.3(DISC-2.3):description, acceptability,prevalencerates,andperformanceintheMECAStudy. JAmAcadChildAdolescPsychiatry 35:865 Y 877 SheldonSH(1998),Pro-convulsanteffectsoforalmelatonininneurolo- gicallydisabledchildren. Lancet 351:1254 SmitsMG,NagtegaalEE,vanderHJ,CoenenAM,KerkhofGA(2001), Melatoninforchronicsleeponsetinsomniainchildren:arandomized placebo-controlledtrial. JChildNeurol 16:86 Y 92 SmitsMG,vanStelH,vanderHeijdenKB,MeijerAM,CoenenAM, KerkhofGA(2003),Melatoninimproveshealthstatusandsleepin childrenwithidiopathicsleeponsetinsomnia.Arandomized placebo-controlledstudy. JAmAcadChildAdolescPsychiatry 42: 1286 Y 1293 SteinMA(1999),Unravellingsleepproblemsintreatedanduntreated childrenwithADHD. JChildAdolescPsychopharmacol 9:157 Y 168 SweereY,KerkhofGA,DeWeerdAW,KamphuisenHA,KempB, SchimsheimerRJ(1998),ThevalidityoftheDutchSleepDisorders Questionnaire(SDQ). JPsychosomRes 45:549 Y 555 TjonPianGiCV,BroerenJP,StarreveldJS,VersteeghFG(2003), Melatoninfortreatmentofsleepingdisordersinchildrenwithattention deficit/hyperactivitydisorder:apreliminaryopenlabelstudy. EurJ Pediatr 162:554 Y 555 VanderHeijdenKB,SmitsMG,VanSomerenEJ,GunningWB(2005a), Predictionofmelatoninefficacybypretreatmentdimlightmelatonin onsetinchildrenwithidiopathicchronicsleeponsetinsomnia. JSleep Res 14:187 Y 194 VanderHeijdenKB,SmitsMG,VanSomerenEJW,GunningWB (2005b),Idiopathicchronicsleeponsetinsomniainattention-deficit/ hyperactivitydisorder:acircadianrhythmsleepdisorder. ChronobiolInt 22:559 Y 570 VanSomerenEJ(2000),Morethanamarker:interactionbetweenthe circadianregulationoftemperatureandsleep,age-relatedchanges,and treatmentpossibilities. ChronobiolInt 17:313 Y 354 VanSomerenEJ,SwaabDF,ColendaCC,CohenW,McCallWV, RosenquistPB(1999),Brightlighttherapy:improvedsensitivitytoits effectsonrest-activityrhythmsinAlzheimerpatientsbyapplicationof nonparametricmethods. ChronobiolInt 16:505 Y 518 VogelsT,VerripsGHW,KoopmanHM,TheunissenNCM,FekkesM, KamphuisRP(2000), TACQOLManual;ParentFormandChildForm . Leiden,TheNetherlands:LeidenChildCenterforChildHealthand PaediatricsLUMC-TNO WaltersAS,MandelbaumDE,LewinDS,KuglerS,EnglandSJ,MillerM (2000),Dopaminergictherapyinchildrenwithrestlesslegs/periodic limbmovementsinsleepandADHD.DopaminergicTherapyStudy Group. PediatrNeurol 22:182 Y 186 WeissMD,WasdellMB,BombenMM,ReaKJ,FreemanRD(2006),Sleep hygieneandmelatonintreatmentforchildrenandadolescentswith ADHDandinitialinsomnia. JAmAcadChildAdolescPsychiatry 45:512 Y 519 ZhdanovaIV,WurtmanRJ,LynchHJ(1995),Sleep-inducingeffectsoflow dosesofmelatoniningestedintheevening. ClinPharmacolTher 57:552 Y 558 MELATONINTREATMENTINADHD 241 J.AM.ACAD.CHILDADOLESC.PSYCHIATRY,46:2,FEBRUARY2007