TheEffectsofCardiorespiratoryBiofeedbackandDialec9calBehavioralSkillsTrainingwithSertralineonPostMyocardialInfarc9onMajorDepressionandLowHeartRateVariability

ADisserta9onPresentedbyPriyaChaudhri,M.A.

CardiovascularDisease

•  CVDistheleadingcauseofmorbidityandmortalityintheU.S.sincethe1940’s(AHA,2005)

•  2,600peopledieofCVDintheU.S.eachday,whichisapproximately1deathevery30seconds(AHA,2005)

•  ThisyearanesMmated1.2millionAmericanswillhaveaneworrecurrentcoronaryaNack(WHO,2006)

CardiovascularDiseaseandDepression

•  ItisesMmatedthat40‐65%ofMIpaMentshavecomorbiddepression(AHA,2005)

•  1outof5MIpaMentshavediagnosableMDDandaquarterhaveminordepression(Carneyetal.,1987)

•  Depressionisassociatedwitha2‐7foldelevatedriskofcardiacevents(Molinarietal.,2006)

•  Post‐MIdepressionisassociatedwitha2‐3foldincreasedriskforcardiacmortality(vanJoostetal.,2004)

•  DepressioncontributestopooradherencetocardiacrehabilitaMon(Kessleretal.,1997)

DepressionandANSDysregula9onDepressionisassociatedwith…•  Elevatedheartrate•  Lowheartratevariability•  Exaggeratedheartrateresponsestophysicalandpsychologicalstressors

•  HighvariabilityinventricularrepolarizaMon

•  LowbaroreceptorsensiMvity•  Elevatedinflammatoryresponse•  Elevatednorepinephrine(NE)whichincreasesSNSacMvity (Carneyetal.,2005)

CVDandHeartRateVariability•  HRVisdefinedasthefluctuaMonsorbeattobeatalteraMonsinHRasmeasuredinms

•  HRVisthesinglegreatestpredictorofmorbidityandmortalityinCVDpaMents(Kleiger,etal.,2000)

•  LowHRVisassociatedwitha2‐4foldincreasedriskofmortalityinpost‐MIpaMents(Biggeretal.,1993)

•  HRVissignificantlylowerindepressedpopulaMons(Carney,etal.,2000)

Limita9onsofCurrentTreatmentOp9ons

•  ThecurrentstandardofcaretotreatdepressionforcardiacpaMentsisSSRImedicaMon,whichhasbeenshowntobeonly40‐60%effecMveinreducingdepressivesymptoms.(Carneyetal.,2005)

•  Betablockershavebeenshowntohavemodesteffectsonincreasingheartratevariability.(Sandroneetal.,1994)

•  AlthoughpharmacologicaltreatmenthasbeenparMallyeffecMvefordepressionandlowHRV,therearemajorconcernswithlongtermconsequences,sideeffects,compliance,andlifeMmecost.

StudyObjec9veToexaminetheefficacyofcardiorespiratorybiofeedbackwithdialec9calbehavioraltherapy(DBT)inconjunc9onwithsertralinemedica9onforthetreatmentofmajordepressivedisorderandlowheartratevariabilityinpost‐MIpa9ents.

DailyPoints

Every9meapointisobtaineditisaddedintheleRhandcorner.Thegoalistoget2or3squarescon9nuouslyinordertoreachthedailygoalof50‐100pts.

Squarestellyouwhetheryouarerelaxed…

3squares=1point

2squares=½point

1square=1point

PulseRateWave

Reflectstheac9vityofthenervoussystemorRSAandrepresentsthespontaneousrhythmicincreaseanddecreaseoftheheartrate.

InfraredFingerSensor

TheStressEraser,Helicor,Inc.

AnindicatorofpoorbaroreceptorfuncMoning,strainedbreathingoremoMonalstress

Synchronizedbreathingatlessthan6.5breathsperminute.

SynchronizedComfortableBreathing

StrainedBreathing

Dialec9calBehavioralTherapy

•  Tradi9onalDBT(1xweekfor2.5hours)•  4ModulesofTreatment:*MindfulnessSkills

*InterpersonalEffec9venessSkills

*DistressToleranceSkills

*Emo9onRegula9onSkills

•  DBTSkillsTrainingforthisstudy(1xweekfor1.5hours)•  3ModulesofTreatment:*MindfulnessSkills

*DistressToleranceSkills

*Emo9onRegula9onSkills

TreatmentProtocol

ExperimentalGroup

CardiorespiratoryBiofeedback20minutedailypracMceofStressEraserdevice

Dialec9calBehavioralSkillsTraining90minuteweeklyDBTgroup

An9depressantMedica9onsertralinetreatment

ControlGroup

An9depressantMedica9onsertralinetreatment

Hypotheses Hypothesis1:ParMcipantsintheexperimentalgroupwouldshowagreaterdecreaseindepressivesymptomsatpost‐treatmentandfollow‐uprelaMvetotheanMdepressantcontrolgroup.

Hypothesis2:Heartratevariability,asmeasuredbySDNN,LF/HFraMo,VLFwouldimprovemoreatpost‐treatmentandfollow‐upfortheexperimentalgroup,thanthecontrolgroup.

Hypothesis3:ParMcipantsintheexperimentalgroupwouldshowgreaterimprovementsinmindfulnessandemoMonregulaMonscoresatpost‐treatmentandfollow‐uprelaMvetotheanMdepressantcontrolgroup.

Hypothesis4:TheexperimentalgroupwoulddemonstrateasignificantcorrelaMonwithStressEraserbiofeedbackpointsandimprovementsindepressionscoresatpost‐treatmentandfollow‐up.

Hypothesis5:Decreasesindepressionscoreswouldbemediatedbyimprovementsinmindfulness,emoMonregulaMonandSDNNatpost‐treatmentandfollow‐up.

DepressionInterviewandStructuredHamilton(DISH)BeckDepressionInventoryII(BDI‐II)Difficul9esinEmo9onRegula9onScale(DERS)FiveFacetMindfulnessQues9onnaire(FFMQ)

Psychological

StandardDevia9onofNormal‐NormalBeats(SDNN)LowFrequency/HighFrequencyRa9o(LF/HF)

VeryLowFrequency(VLF)

HeartRateVariability

Medica9onLogBreathingLog

WeeklyLogs

AssessmentPeriodsBaseline(Week1)Post‐Treatment(Week8)Follow‐Up(Week12)

Measurements

RS

B RS

B RPBSPB

B

PRE POST Follow‐Up1 2 3 4 5 6 7 8 9 10 11

B=baseline;S=stressor;R=recovery;PB=pacedbreathing

PhysiologicalMeasurementofHRV

Baseline:10min(listenedtotravellog)

Stressor:2minuteserial7test

Recovery:3minuterecovery(siingquietly)

PacedBreathing:5minof6BPMbreathing

• DocumentedCVDandMI. • Absenceofrecentcardiacevents(MI<2months)andmedicaMonschangesinthe4weekspriortothefirsttreatmentvisit. • AprimarydiagnosisofMajorDepressiveDisorder,asassessedbyaDISHHamiltonscoreof18orgreater,withlowriskofsuicidedefinedbyaBDIscoreoflessthan2onquesMon#9. • StabilizedonsertralinemedicaMon(atleast4weeks)priortostudy. • StablecogniMvefuncMoningbasedontheMini‐MentalStatusExam(MMSE).

• Notbetweentheagesof20‐85,metthecriteriaforClassIVCongesMveHeartFailure,hadapacemaker,and/orpregnant. • ExceededadailyconsumpMonof830milligramsofcaffeine. • PresenceofanyphysicalcondiMonsormedicaMonsthatmakeheartratevariabilityuninterrupMble. • CoronaryintervenMonwithinthepast2months. • SeriouscomorbidmedicalcondiMonthatmayaffectdepressionlevels(i.e.hypothyroidism,diabetes). • PracMcedweeklyyogaormeditaMon.

InclusionCriteria ExclusionCriteria

Demographics

•  60TotalPar9cipants•  34Males•  26Females•  55=MeanAge•  50%Caucasian•  70%Married•  70%Collegeeduca9on

PrimaryAnalyses

Correla9onAnalysesPearsoncorrelaMontests.

StructuralequaMonmodeling(SEM),SobelmediaMontest.Media9onAnalyses

Sta9s9calAnalyses

MulM‐levelModelingwasimplementedusinghierarchicallinearmodeling(HLM).ArandominterceptHLMmodelwasusedtoanalyzethepsychologicalmeasures(HAM‐D,BDI‐II,DERS,FFMQ).AnunstructuredHLMmodelusingrandominterceptandrandomslopewasusedtoanalyzeheartratevariability(SDNN),(LF/HFraMo),(VLF)acrossthreemeasurementcondiMons(baseline,stressor,andrecovery).ToassessthepotenMalimpactofmissingdata,arandomeffectspaNern‐mixtureanalysiswasimplementedwithabinarymissingdatavariable(allparMcipantsversuscompleters)whichwasenteredasapredictorintherandomregressionmodel.

ResultsoftheStudy

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

Pre Post Follow‐up

MeanHam

ilton

HamiltonScoresAcrossTime(N=60)

biofeedbackcontrol

HLMAnalysisofHamiltonScoresAcrossTime

AllPar9cipants(N=60)

Est. SE p<

TimexTreatment ‐3.20 1.12 .006 **DropoutxTimexTreatment .591

*(p<.05)

**(p<0.01)

***(p<.001)

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

Pre Post Follow‐up

MeanBD

I

biofeedbackcontrol

BDIScoresAcrossTime(N=60)

HLMAnalysisofBDIScoresAcrossTime

AllPar9cipants(N=60)

Est. SE p<

TimexTreatment ‐8 1.6 <.0001***DropoutxTimexTreatment 0.002 *

*(p<.05)

**(p<0.01)

***(p<.001)

DISHDepressionDiagnosisFrequencyDistribu9onatPost‐Treatment

no major / no minor dep

DISHDepressionDiagnosisFrequencyDistribu9onatFollow-Up

no major / no minor dep

MajorDepressiveDisorderAcrossTime(N=60)

0

2

4

6

8

10

12

Pre Post Follow-Up

Maj

or D

epre

ssio

n (M

ean)

Time

Major Depression Diagnosis

Exp Control

SDNN Est. SE p<Stressor

TimeXTreatment 0.009 0.005 0.055 γ DropoutXTimeXTreatment 0.134

RecoveryTimeXTreatment 12.696 4.555 0.004**DropoutXTimeXTreatment 0.138

LF/HFRa9oStressor

TimeXTreatment 0.051 0.403 0.213DropoutXTimeXTreatment 0.138

RecoveryTimeXTreatment 0.362 0.479 0.453DropoutXTimeXTreatment 0.590

VLFStressor

TimeXTreatment 0.374 2.441 0.879DropoutXTimeXTreatment 0.144

RecoveryTimeXTreatment 8.383 2.437 0.001*** DropoutXTimeXTreatment 0.130

HLMAnalysisofHeartRateVariability

*significant(p<0.05);**significant(p<0.01);***significant(p<0.0001);γ sta9s9caltrend

00.010.020.030.040.050.06

Pre‐Treatment Post‐Treatment Follow‐Up

ExpControl

StandardDevia9onofNormal‐NormalBeats(SDNN)MeanAcrossAllMeasurementCondi9ons(Baseline,Stressor,Recovery)

0

5

10

15

20

25

Pre‐Treatment Post‐Treatment Follow‐Up

ExpControl

VeryLowFrequency(VLF)MeanAcrossAllMeasurementCondi9ons(Baseline,Stressor,Recovery)

6.00

7.00

8.00

9.00

10.00

11.00

12.00

13.00

14.00

Pre Post Follow‐up

MeanFFMQ

biofeedback

control

FFMQScoresAcrossTime(N=60)

HLMAnalysisofFFMQScoresAcrossTime

AllPar9cipants(N=60)

Est. SE p<

TimexTreatment 0.28 0.05 <.0001 ***

DropoutxTimexTreatment .132

*(p<.05)

**(p<0.01)

***(p<.001)

0.00

20.00

40.00

60.00

80.00

100.00

120.00

140.00

Pre Post Follow‐up

MeanDER

S

DERSScoresAcrossTime(N=60)

biofeedback

control

HLMAnalysisofDERSScoresAcrossTime

AllPar9cipants(N=60)

Est. SE p<

TimexTreatment 28.00 3.68 <.0001 ***

DropoutxTimexTreatment .150

*(p<.05)

**(p<0.01)

***(p<.001)

PearsonCorrela9on

pvalue n

BDI‐0.371 .052γ 28

Post‐TreatmentBDI

‐0.535 .022* 18Follow‐UpHamilton

‐0.349 .069γ 28Post‐Treatment

Hamilton‐0.476 .040* 18

Follow‐Up

*(p<.05)

γ (p<.10),sta9s9caltrend

StressEraserPointsCorrela9onwithDepressionOutcomes

Ham

ilton

ScoresatPost‐Treatm

ent

BDIScoresatPost‐Treatm

ent

StressEraserPointsCorrela9onswithDepressionOutcomesatPost‐treatment

Condi9on

DERS

Hamilton

Media9onPathAnalysisModel

Unstandardized Standard ZScoreRegressionCoefficient Error

Post‐Treatment(N=53)

DERSonHamilton ‐1.32 0.74 ‐1.80DERSonBDI ‐1.53 0.44 ‐3.46*FFMQonHamilton 0.72 0.4 1.81FFMQonBDI 0.20 0.23 0.85SDNNonHamilton ‐0.27 0.37 ‐0.74SDNNonBDI ‐1.48 0.38 ‐3.92*

Follow‐Up(N=30)DERSonHamilton ‐4.33 1.11 3.90*DERSonBDI ‐5.35 1.52 ‐3.53*FFMQonHamilton 1.31 0.65 2.01*FFMQonBDI ‐0.91 0.64 ‐1.42SDNNonHamilton ‐1.33 0.68 ‐2.07*SDNNonBDI ‐1.46 0.36 ‐3.90*

*Significanceis(Z>1.96)or(Z<1.96)

MediatorstoImprovementsinDepressionOutcomes

SummaryofDescrip9veDataforallOutcomeVariables

Pre‐Treatment(n=60)

Post‐Treatment(n=53)

Follow‐Up(n=30) Slope EffectSize

M (SD) M (SD) M (SD) Post‐Treatment

HamiltonExp 29.2 (4.47) 14.14 (5.05) 12.94 (3.64) ‐8.9*** 1.4

Con 27.67 (4.64) 18.64 (4.92) 17.75 (4.67) ‐5.7*** 1.0

BDI Exp 30.95 (8.69) 8.89 (7.71) 7.5 (6.99) ‐12.6*** 1.5

Con 30.53 (6.33) 24.56 (5.56) 21.95 (6.64) ‐4.5*** 0.8

FFMQ Exp 10.13 (1.96) 12.58 (1.92) 13.23 (1.23) 0.32*** 1.3

Con 10.81 (1.7) 11.01 (1.76) 11.63 (1.26) 0.04 0.4

DERS Exp 123.13 (16.54) 68.96 (20.42) 64.17 (15.00) 31.7*** 1.5

Con 103.47 (19.92) 98.04 (19.01) 93.67 (15.14) ‐3.7 0.5

SDNNAcrossphases

Exp 0.042 (0.012) 0.057 (0.020) 0.048 (0.016) 1.30** ‐0.9

Con 0.031 (0.020) 0.049 (0.021) 0.035 (0.019) 0.0046 ‐0.9

LF/HFAcrossPhases Exp 2.39 (1.79) 1.89 (0.95) 1.21 (0.71) ‐0.225 0.4

Con 1.86 (1.03) 2.32 (1.26) 1.94 (1.32) 0.297 ‐0.4

VLFAcrossPhases

Exp 19.62 (9.21) 8.98 (4.13) 7.03 (8.63) ‐8.40*** 1.6Con 11.28 (10.88) 14.01 (7.84) 14.14 (11.99) ‐0.303 ‐0.3

*significant(p<0.05);**significant(p<0.01);***significant(p<0.0001);γ sta9s9caltrend

TimexTreatmentInterac9onEffects

Measures SlopeDifferences pValue EffectSize

Hamilton ‐3.20 .006 ** 1.0

BDI ‐8.00 <.0001 *** 2.0

FFMQ 0.28 <.0001 *** 0.9

DERS 28.00 <.0001 *** 1.6

SDNN 0.6Baseline ‐0.003 .848

……..Stressor 0.0093 .055 γ Recovery ‐0.0019 .004 **

LF/HFRa9o 0.3

Baseline 0.4704 .316……..Stressor 0.0508 .213

Recovery 0.3616 .453

VLF 0.7

Baseline 8.4248 .013 *………Stressor 0.3744 .879

Recovery 8.3827 .001 **

*significant(p<0.05);**significant(p<0.01);***significant(p<0.0001);γ sta9s9caltrend

Conclusions

•  ThisstudyrevealedthatthecardiorespiratorybiofeedbackandDBTorientedskillstraininginconjuncMonwithsertralineresultedinasignificantlylargerimprovementindepressionseverity,depressiondiagnosis,andheartratevariabilityatpost‐treatmentcomparedtothecontrolgroup,andthiseffectwasmaintainedatfollow‐up.InaddiMon,theexperimentalgroupshowedasignificantlylargerimprovementinemoMonregulaMonandmindfulnessscoresacrossMme.

WhyDidTheExperimentalGroupHaveBererOutcomes?

Post‐Treatment(N=53)

Follow‐Up(N=30)

CategoriesExperimental

GroupControlGroup

ExperimentalGroup

ControlGroup

TimeConflict 2 1 7 4

SSRIDiscon9nued 0 1 4 5

Disinterested 0 1 0 4

CardiacReoccurrence 0 2 1 5

n=7 n=30

ReasonsForDropOut

Theresultsofthisstudycannotbeexplainedbythefollowingconfoundsatbaseline:

• Depressiondiagnosisandseverity• Medica9ons• Gender• Age• Ethnicity• Socioeconomicstatus• Par9cipants’treatmentexpectancies

AlthoughthepaNernsofdropoutmayhavebiasedtheresultsfortheBDIatfollow‐up,therewasnoevidencethattheHamilton,heartratevariability,emoMonregulaMon,andmindfulnessresultswereaffectedbydropouts.

GroupEquivalence

BererOutcomesMayBeArributedTo…..

DBTSkills*EmoMonregulaMon*Distresstolerance*Mindfulness*BehavioralacMvaMon

ImprovedHeartRateVariability*Cardiorespiratorybiofeedback*Breathingretraining

Self‐Efficacy*Empowerment

ClinicalImplicaMons•  DecreasesindepressionandincreasesinHRV

canimproveANSregulaMon‐‐‐‐whichmayreduceriskofcardiacmorbidityandmortality

•  DecreasesinphysiologicalreacMvitytostressorsandimprovedrecoveryfromitmayalsoprotectagainstthedevelopmentorworseningofCVD

•  BeNeradherencetocardiacrehabilitaMon•  MoreaNenMvetobehavioralriskfactors

(tobacco/alcoholuse,physicalinacMvity,poordiet)

•  ImprovedemoMonalandphysicalhealth•  Improvedqualityoflifeandself‐efficacy•  ImprovedmindfulnessandemoMon

regulaMonmaypreventrelapserateswithfuturedepressiveepisodes

•  ItisnoteworthythatthecontrolgrouphadmoredropoutsaNributedtocardiacreoccurrence.

Acknowledgements

ResearchAssistantsAmyMcKinneyChrisMnaHuangMishanehMarjani

CommiRee

Dr.MiltonBrown,Ph.D.Dr.RichardGevirtz,Ph.D.

Dr.RayGandhi,M.D.Dr.SharonFoster,Ph.D.

Aspecialthanksto…

Dr.FredMuench,Ph.D.Dr.DaleGlaser,Ph.D.

PostHocMedia9onwithIndividualMindfulnessFacetstoImprovementsinDepressionOutcomes

ion pValues

Hamilton BDI

IndividualFacetsofFFMQ

observing .163 .159

describing .124 .107

acMngwithawareness .077γ .045*

non‐judgingofinnerexperience .025* .028*

non‐reacMvitytoinnerexperience .070γ .049*

*significant(p<0.05);γ sta9s9caltrend

pValues

IndividualSubscalesofDERS Hamilton BDI

non‐acceptanceofemoMonalresponses 0.837 0.984

difficulMesengagingingoaldirectedbehavior 0.590 0.571

impulsecontroldifficulMes 0.509 0.936

lackofemoMonalawareness 0.918 0.805

limitedaccesstoemoMonregulaMonstrategies 0.678 0.517

lackofemoMonalclarity 0.990 0.316

PostHocMedia9onwithIndividualSubscalesforDERStoImprovementsinDepressionOutcomes