Depression is associated with increased vagal withdrawal during unpleasant emotional imagery after cardiac surgery

Page 1

Autonomic Neuroscience: Basic and Clinical xxx (2015) xxx–xxx

AUTNEU-01733; No of Pages 8

Contents lists available at ScienceDirect

Autonomic Neuroscience: Basic and Clinical

j ourna l homepage: www.e lsev ie r .com/ locate /autneu

Depression is associated with increased vagal withdrawal duringunpleasant emotional imagery after cardiac surgery

Elisabetta Patron a,⁎, Simone Messerotti Benvenuti a, Giuseppe Favretto b,Renata Gasparotto b, Daniela Palomba a

a Department of General Psychology, University of Padova, Italyb Department of Preventive and Rehabilitative Cardiology, Motta di Livenza Hospital, Italy

⁎ Corresponding author at: Department of General PsyVia Venezia, 8, 35131 Padova, Italy. Tel.: +39 049 827 66

E-mail address: [email protected] (E

http://dx.doi.org/10.1016/j.autneu.2015.02.0021566-0702/© 2015 Elsevier B.V. All rights reserved.

Please cite this article as: Patron, E., et al., Decardiac surgery, Auton. Neurosci. (2015), htt

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 6 February 2014Received in revised form 25 November 2014Accepted 2 February 2015Available online xxxx

Keywords:Cardiac surgeryCardiovascular reactivityDepressionHeart rate variabilityImageryMood

Background: The aim of this study was to examine the influence of depression on heart rate and heart ratevariability (HRV) during emotional imagery in patients after cardiac surgery.Methods: Based on the scores of the Center for Epidemiological Studies of Depression (CES-D) scale, 28 patientsafter cardiac surgery were assigned either to the group with depression (CES-D scores≥ 16; N=14) or the onewithout depression (CES-D scores b 16;N=14). Each patient completed a rest period and an emotional imageryincluding pleasant, neutral and unpleasant scripts. Inter-beat intervals (IBIs) and HRVweremeasured during theentire protocol.Results: Compared to nondepressed patients, those with depression had greater reductions in high frequencyexpressed in normalized units (HF n.u.) during the imaging of the unpleasant script (p = .003, Cohen's d =1.34). Moreover, HF n.u. were lower during the imaging of the unpleasant script than the pleasant one indepressed patients only (p = .020, Cohen's d = 0.55). CES-D scores were also inversely correlated withresidualized changes in IBIs (r = − .38, p = .045) and HF n.u. (r = − .49, p = .008) from rest to the imaging

of the unpleasant script.Conclusions: The relationship between depression and increased vagal withdrawal during unpleasant emotionalimagery extends to patients after cardiac surgery. The present study suggests that increased vagal withdrawal tonegative emotions in patients after cardiac surgery may mediate the conferral of cardiac risk by depression.

© 2015 Elsevier B.V. All rights reserved.

1. Introduction

Depression is an important and independent risk factor for the onset,the adverse course and outcomes in patients with coronary arterydisease (CAD) (Carney et al., 2005;Wulsin and Singal, 2003). Specifical-ly, there is evidence that depression is associated with a 60% greaterlikelihood of having CAD (Wulsin and Singal, 2003). Patients withdepression are also more likely to have a major cardiac event within ayear of the diagnosis of CAD (Carney et al., 1988) and/or to die in theyears following the diagnosis (Barefoot et al., 1996). Moreover, it hasbeen consistently reported that depression is a risk factor for cardiacmorbidity and/or mortality in patients who had undergone cardiacsurgery (Blumenthal et al., 2003; Lespérance et al., 2000).

The autonomic nervous system (ANS) has been identified as apivotal site of dysregulation, with reduced parasympathetic and/orincreased sympathetic nervous system activity leading to arrhythmiasand sudden cardiac death (Musselman et al., 1998; Podrid et al., 1990;

chology, University of Padova,35; fax: +39 049 827 6600.. Patron).

pression is associated with inp://dx.doi.org/10.1016/j.autn

Schwartz and Vanoli, 1981). Given that depression itself is associatedwith reduced parasympathetic and/or increased sympathetic activity(Kemp et al., 2010), the presence of depression may potentiate theimpaired parasympathetic control and increased sympathetic activityobserved in patients with CAD. In turn, these depression-related ANSabnormalities, especially a reduced cardiac vagal tone, may furtherpredispose depressed patients with CAD to ventricular tachycardia,ventricular fibrillation, myocardial ischemia, and sudden cardiac death(Carney et al., 2002, 2005).

In line with these findings, there is abundant and convergingevidence that, compared to nondepressed individuals, depressedpatients with CAD or after myocardial infarction have reduced heartrate variability (HRV), which is associated with excessive sympatheticmodulation and/or inadequate cardiac vagal control (Carney et al.,2001, 2002, 2005). Intriguingly, it has been recently shown that adepression-reduced HRV association extends also to patients whounderwent cardiac surgery (Patron et al., 2012, 2014). In particular,Patron et al. (2012, 2014) observed that, compared to patients withoutdepression, those with depression have reduced HRV, especially animpaired cardiac vagal tone, at discharge from the hospital, thussuggesting an association between postoperative depression and

creased vagal withdrawal during unpleasant emotional imagery aftereu.2015.02.002

Page 2

2 E. Patron et al. / Autonomic Neuroscience: Basic and Clinical xxx (2015) xxx–xxx

cardiac vagal dysfunctions after cardiac surgery. Based on this finding,reduced HRV has been suggested as one possible pathophysiologicalmarker of depression-related ANS dysregulation that has been implicat-ed as a risk factor for cardiacmorbidity and/ormortality in post-surgicalpatients (Blumenthal et al., 2003; Lespérance et al., 2000).

In addition to the depression–ANS dysregulation relationship at rest,some studies have shown depression-related abnormalities in vagalactivity during psychophysiological stressors such as the speech taskor the cold pressor task, indicating reduced ability to regulate cardiacactivity to meet task demands (Appelhans and Luecken, 2006; Hughesand Stoney, 2000; Sheffield et al., 1998). Specifically, it has been report-ed that, compared to individuals without depression, CAD or healthyindividuals with depression are characterized by greater decreases invagal activity, as reflected by reduced high frequency (HF) power ofHRV, during a speech task — a stressor that generally elicits reductionin HF power (Berntson et al., 1994; Hughes and Stoney, 2000; Sheffieldet al., 1998). Consistent with these studies, a recent meta-analysisreported that depression is associated with increased heart rate reactiv-ity, especially in those patients with cardiovascular diseases (Kibler andMa, 2004). In particular, Kibler and Ma (2004) suggested that thedepression-related autonomic imbalance may predispose depressedpatients with cardiovascular diseases to exaggerated physiologicalstress reactions and therefore increase cardiac risk.

By contrast, however, a growing body of research has shown thatdepression may be associated with attenuated rather than increasedcardiovascular reactivity in healthy individuals (e.g., Carroll et al.,2007; Schwerdtfeger and Rosenkaimer, 2011), as well as in patientswith major depression disorder (e.g., Salomon et al., 2009), and inpatients with CAD (York et al., 2007). In line with these findings,it should be noted that the effect sizes of the relationship betweendepression and exaggerated cardiovascular reactivity reported byKibler and Ma (2004) were only small to moderate. The latter may bedue to the fact that the vast majority of studies examining whetherand how cardiac vagal activity could be modulated by depression usednon-emotional tasks (e.g., mental arithmetic, cold pressor or Strooptask) (Kibler and Ma, 2004). Conversely, the number of studies thatinvestigated the influence of depression on cardiac vagal activity duringemotional stressors is muchmore limited. This is surprising given that itis well-established that depression is characterized by a bias towardnegative affect or emotion, namely the mood-congruent bias(Eizenman et al., 2003; Erickson et al., 2005; Mogg and Bradley, 1998;Murphy et al., 1999), which, in turn, has been found to potentiateemotional and psychophysiological reactivity to negative stimuli orcontext (Golin et al., 1977; Lewinsohn et al., 1973).

In order to examine the influence of depression on cardiac vagalactivity during an emotional task in patients who have undergonecardiac surgery, a useful tool is the emotional imagery paradigm.Specifically, the emotional imagery task can effectively be used as anemotional stressor in order to induce and/or manipulate emotionalstates (e.g., happiness, fear), and related physiological responses(e.g., Lang, 1979; Lang et al., 1990). That is, the emotional imagerytask can influence both the intensity of the affective experience andthe physiology associated with the affective experience during mentalimagery (Lang, 1979). For example, at the subjective level, the imageryof stress-evoking vs. neutral scripts has been shown to be effective inincreasing symptoms associated with anxiety (e.g., Williams andCumming, 2012; Williams et al., 2010). At the physiological level,emotional imagery has been found to elicit somatic and autonomicresponses that are isomorphic with those typically reported in thereferenced behavioral context (Cuthbert et al., 1991; Lang et al., 1990).Among autonomic responses, there is considerable evidence that imag-ery elicits significant cardiac acceleration in response to emotionallyarousing than neutral scripts (e.g., Gollnisch and Averill, 1993; Langet al., 1990; vanOyen Witvliet and Vrana, 1995; Vrana et al., 1988;Williams et al., 2010), reflecting the activation of the approach- andwithdrawal-related behaviors (Bradley, 2000).

Please cite this article as: Patron, E., et al., Depression is associated with incardiac surgery, Auton. Neurosci. (2015), http://dx.doi.org/10.1016/j.autn

In light of these considerations, the present study was designed toexamine whether postoperative depression would be associated withincreased vagal withdrawal in patients after cardiac surgery during anemotional imagery task, including pleasant, neutral and unpleasantconditions. Given that depression is characterized by increased psycho-physiological reactivity to negative stimuli, post-surgical patients withdepression were expected to exhibit greater heart rate increase andvagal withdrawal, as reflected by lower HF power of HRV, during theimaging of the unpleasant vs. the neutral and the pleasant scriptscompared to patients without depression.

2. Materials and methods

2.1. Participants

After receiving the local ethics committee's approval, 28 patientswho had undergone first-time cardiac surgery were enrolled in thisstudy after their written informed consent was obtained. Based on thepresence of clinically significant depression [i.e., scoring greater than16 on the Center for Epidemiologic Studies of Depression (CES-D)scale] (Radloff, 1977; Italian version by Fava, 1982), the patients wereclassified into one of two groups: with depression (N= 14) or withoutdepression (N = 14). CES-D is a 20-item, self-reporting questionnairedesigned to measure the presence of common symptoms of depressionover the previous week. CES-D score ranges from 0 to 60, with thehigher scores indicating greater depressive symptoms. Each patientwas administered the validated Italian version of the CES-D question-naire, which has shown good psychometric properties (Fava, 1982).All patients had undergone cardiac surgery at a regional specializedhospital and were admitted for rehabilitation in a highly specializedhospital between July 2012 and December 2012. Patients underwentheart valve surgery (N = 12), coronary artery bypass graft (CABG)surgery (N = 12), and combined surgery (heart valve plus CABG sur-gery) (N = 4). Each patient had the same protocol of cardioplegia anda mild hypothermic cardiopulmonary bypass. Ages greater than 75,incapability to read or understand Italian language, visual or auditoryimpairments, use of psychotropic drugs, other life-threatening medicalillness, and prior cerebrovascular and/or neurological diseases werethe exclusion criteria.

2.2. Imagery task

Three narratives, selected from the Affective Norms of English Text(ANET) (Bradley and Lang, 2007) based on standardized ratings ofvalence and arousal, were translated and adapted for the Italian popula-tion while maintaining the meaning and the intention of the originalitems, and categorized as pleasant (narrative 8500: Valence, M = 8.5,SD=1.5; Arousal,M=8.2, SD=1.7), neutral (narrative 2610: Valence,M=5.4, SD=1.3; Arousal,M=3.1, SD=1.7) or unpleasant (narrative6020: Valence, M = 1.9, SD = 1.2; Arousal, M = 8.2, SD = 1.5).The scripts from the English original and Italian translation and adapta-tion are presented in Appendix A. All patients were instructed to listento the script read by the same experimenter and upon offset vividlyimagine themselves in the event described. The experimenter read thescripts slowly (for a total duration of 30 s each), in order for the patientsto fully understand each script. Each imagery trial consisted of 2 min ofactive imagery. After each imagery trial, in order to ascertain that theparticipants imagined the specific scene, the experimenter asked themto describe the details of the scene they imagined and the emotionsexperienced during the imagery. Then, valence and arousal of theaffective state associated with the scene imagined were rated usingthe Self-Assessment Manikin (SAM) scale (Bradley and Lang, 1994).On both the valence and the arousal dimensions, the SAM scoresrange from 1 (valence: very unpleasant; arousal: very calm) to 9(valence: very pleasant; arousal: very aroused). After each descriptionand self-reported assessment of valence and arousal of the affective

creased vagal withdrawal during unpleasant emotional imagery aftereu.2015.02.002

Page 3

3E. Patron et al. / Autonomic Neuroscience: Basic and Clinical xxx (2015) xxx–xxx

state associatedwith the imagined scene, 2min of restwas allowed. Theorder of the presentation of the three narratives was counterbalancedacross participants.

2.3. HRV recordings

HRV was recorded in a standardized fashion using a computerizedrecording system (ProComp Infiniti, Thought Technology; Montreal,Canada). The electrocardiographic (ECG) signal was obtained fromthree disposable Ag/AgCl electrodes that were positioned on thepatient's chest in a modified lead II configuration. Each ECG signal wasamplified, band-pass filtered (1–100 Hz), and sampled at 256 Hz. Adigital trigger detecting R-waves was applied to ECG signal to obtaininter-beat intervals (IBIs). All ECG data were visually inspected andartifacts were corrected with a piecewise cubic spline interpolationmethod that generates missing or corrupted values into the normal-to-normal intervals (or inter-beat intervals). Kubios HRV AnalysisSoftware 2.0 (The Biomedical Signal Analysis Group, Department ofApplied Physics, University of Kuopio, Finland) was then used tocompute the frequency domain indices by applying the fast-Fourier-transform-based Welch's periodogram method (window width =256 s; overlap = 50%) on the normal-to-normal series. The frequencydomain indices were calculated as follows:

1) Low frequency (LF) power (0.04 to 0.15 Hz) in ms2, which reflectsboth sympathetic and parasympathetic cardiac activity and isstrongly related to blood pressure regulation (Kamath et al., 1987).

2) High frequency (HF) power (0.15 to 0.40 Hz) in ms2, which primar-ily reflects cardiac parasympathetic tone (Kamath et al., 1987; TaskForce, 1996).

All frequency domain indices were logarithmically transformed tonormalize their distribution. In addition, LF and HF components wereexpressed in normalized units (n.u.), which represent the relativevalue of each power component in proportion to the total powerminus the very low frequency (VLF) component (0 to 0.04 Hz), andrange from 0 to 100%. The relation between LF and HF components inn.u. emphasizes the controlled and balanced behavior of the twobranches of the ANS, thus accurately reflecting the sympathovagalbalance (Malliani et al., 1994; Task Force, 1996). Moreover, the normal-ization has the advantage to minimize the effect of the changes in totalpower on the values of LF and HF components (Task Force, 1996). Giventhat LF and HF n.u. are algebraically redundant we reported only the HFn.u. in the Results section.

2.4. Procedure

All patients completed the experimental protocol after theiradmission for rehabilitation, approximately two weeks (range 10–14 days) after cardiac surgery. Upon arrival for the session, the patientscompleted the Questionnaire upon Mental Imagery (QMI) (Sheehan,1967), a 35-item self-report questionnaire that measures individualdifferences in imagery skills. QMI requires participants to indicate on7-point scales how lively and vividly they can imagine each item; scoresrange from 35 to 245, with lower scores reflecting better imageryability. Then, after skin preparation, surface electrodes were placedover the right subclavicular space and fifth left intercostal space formeasurement of ECG. After the electrodes were attached, patients rest-ed for 5-min and then completed the imagery task. ECG was recordedduring both the 5-min resting period and the imagery task. In order toavoid movement artifacts, each patient was instructed to stay still andnot to talk during the ECG recordings. All ECG recordings were takenwith the patients seated on a semireclined armchair after adaptationto the laboratory. Each physiological measurement was taken in aquiet and isolated laboratory during morning hours (from 10:00to 12:00).

Please cite this article as: Patron, E., et al., Depression is associated with incardiac surgery, Auton. Neurosci. (2015), http://dx.doi.org/10.1016/j.autn

2.5. Data analysis

As our first step, analysis of variance (ANOVA), with Group (withdepression, without depression) as the between-subjects factor, wasused to compare the age of the two groups. Chi-square or Fisher'sexact test analysis was conducted to compare the two groups in termsof gender, smoking, biomedical risk factors, cardiac drugs, and surgicalprocedures (according to the complexity of cardiac surgery, that is,heart valve surgery or CABG vs. combined surgery).

As our second step, mixed analyses of covariance (ANCOVAs), withGroup (with depression, without depression) as the between-subjectsfactor, and emotional Condition (pleasant, neutral, unpleasant scripts)as a within-subjects factor, and QMI scores as covariate were conductedon self-reported evaluations of valence and arousal.

As our third step, mixed ANCOVAs with Group (with depression,without depression) as the between-subjects factor, emotional Condi-tion (rest, pleasant, neutral, unpleasant scripts) as a within-subjectsfactor, and QMI scores as covariate were conducted on HRV parameters(IBIs, lnLF, lnHF, and HF n.u.) collected during the imagery task. QMIscorewas entered as covariate given that depressed patients had higherQMI scores (i.e., worse imagery ability;M=93.1; SD=34.2) than thosewithout depression (M=72.9; SD=42.6). Indeed, althoughwe did notfind a significant effect of group, when we considered the effect size ofthemean difference between groups in QMI scores, we found amediumeffect size (Cohen's d = 0.52). The significance of main effects andinteractions was adjusted where appropriate using the Greenhouse–Geisser method to correct for violations of sphericity. In the results,the corrected p levels and epsilon (ε) are reported together with theuncorrected degrees of freedom. The Fisher's LSD test was used forpost hoc analyses. Cohen's d was calculated as a measure of the effectsize. The Cohen's d values considered to represent small, medium, andlarge effects are 0.20, 0.50, and 0.80, respectively (Cohen, 1988).

Pearson's correlation coefficients were calculated between CES-Dscores and residualized changes for IBIs and HRV parameters from restto pleasant, neutral and unpleasant imagery conditions. Residualizedchange scores were calculated by regressing IBIs and HRV parametersduring rest on their values during the imaging of pleasant, neutraland unpleasant scripts and calculating the residual of the resultantregression (Messerotti Benvenuti et al., 2012a, 2012b, 2013; Patronet al., 2013a). Therefore, residualized change is an index of changethat takes into account the patient's IBIs and HRV parameters duringrest. A p value of b .05 was considered to be statistically significant.

3. Results

3.1. Characteristics of patients with depression and without depression

Fisher's exact test or chi-square analysis revealed no group differ-ences for gender, smoking, cardiac surgery, beta-blockers, angiotensin-converting enzyme inhibitors, antiarrhythmics, and anticoagulants,hypertension, diabetes, dyslipidemia, and previous myocardial infarc-tion. Similarly, ANOVAyielded nogroupdifferences for age. Thedescrip-tive statistics for each group are reported in Table 1.

3.2. Emotional self-reports

In the case of valence, a significant Group × Condition interaction(F[2, 50] = 5.08, p b .02, ε = .96, η2p = .17) was found. While bothgroups did not differ in terms of self-reported valence ratings for thepleasant and the unpleasant conditions, patients with depression werecharacterized by significantly lower valence ratings for the neutralscript compared to patients without depression (p b .009, Cohen'sd = 0.90). In addition, both groups showed greater affective valenceratings for the pleasant and theneutral scripts compared to theunpleas-ant one (pleasant vs. unpleasant: ps b .001, Cohen's ds ≥ 1.68; neutralvs. unpleasant: ps b .003, Cohen's ds ≥ 1.06). No significant effect for

creased vagal withdrawal during unpleasant emotional imagery aftereu.2015.02.002

Page 4

Table 1Demographic and biomedical characteristics by depression status.

Variable Group with depression(N = 14)

Group without depression(N = 14)

p

Age (years) 56.6 (8.4) 57.0 (8.4) .89Male sex (N, %) 11 (79) 13 (93) .60Diabetes (N, %) 1 (7) 2 (14) .99Beta-blockers (N, %) 11 (79) 12 (86) .99ACE inhibitors (N, %) 7 (50) 5 (36) .70Antiarrhythmics (N, %) 4 (29) 2 (14) .65Anticoagulants (N, %) 13 (93) 14 (100) .99Hypertension (N, %) 8 (57) 5 (36) .45Myocardial infarction(N, %)

0 (0) 2 (14) .48

Dyslipidemia (N, %) 7 (50) 5 (36) .70Cardiac surgery .10

Heart valve or CABG(N, %)

10 (71) 14 (100)

Combined (N, %) 4 (29) 0 (0)Smoking .14

Actual (N, %) 2 (14) 0 (0)Past (N, %) 9 (64) 7 (50)No (N, %) 3 (22) 7 (50)

Notes: Data are M (SD) of continuous and N (%) of categorical variables. ACE =angiotensin-converting enzyme; CABG surgery = coronary artery bypass graft surgery.

4 E. Patron et al. / Autonomic Neuroscience: Basic and Clinical xxx (2015) xxx–xxx

Group (F[1, 25]=3.08, p=.09, η2p=.11) or Condition (F[2, 50]=2.32,p = .11, ε = .96, η2p = .08) was noted.

In the case of arousal, a marginally significant Condition main effect(F[2, 50]= 3.06, p= .07, ε= .84, η2p = .11) was noted. Greater arousalscores were observed for the pleasant and the unpleasant scriptscompared to the neutral one (pleasant vs. neutral: Cohen's d = 0.91;unpleasant vs. neutral: Cohen's d = 1.24). No significant effect forGroup (F[1, 25] = 1.42, p = .24, η2p = .05) or Group × Conditioninteraction (F[2, 50] = 0.37, p = .66, ε = .84, η2p = .01) was noted.The descriptive statistics of self-reported valence and arousal ratingsfor each emotional condition and group are reported in Table 2.

3.3. Effects of depression on HRV during emotional imagery

The Group × Condition ANCOVA on IBIs did not show a significanteffect for Group (F[1, 25] = 0.40, p = .53, η2p = .02), Condition (F[3,75]=0.46, p=.63, ε=.65, η2p=.02) or Group× Condition interaction(F[3, 75] = 1.27, p= .29, ε= .65, η2p = .05). In the case of HRV param-eters, the Group × Condition ANCOVA on HF n.u. yielded a significantmain effect for Group (F[1, 25] = 4.67, p b .05, η2p = .16), and aGroup × Condition interaction effect (F[3, 75] = 3.49, p b .04, ε = .85,η2p=.12). Fisher's LSD post hoc test revealed that, compared to patientswithout depression, those with depression had significantly lower HFn.u. during the imaging of the unpleasant script (p b .01, Cohen's d =1.34), but not at rest (p= .12, Cohen's d= 0.56), or during the imagingof the pleasant (p = .59, Cohen's d = 0.20) and the neutral (p = .08,Cohen's d = 0.67) scripts. Moreover, in patients with depression, asignificantly lower HF n.u. during the imaging of the unpleasant scriptcompared to the pleasant one (p b .03, Cohen's d = 0.55), but not for

Table 2Ratings of each self-report measure in the group with and without depression.

Self-reportmeasure

Group with depression (N = 14) Group without depression(N = 14)

Pleasant Neutral Unpleasant Pleasant Neutral Unpleasant

Valence 7.1 (2.3) 5.8 (2.7) 3.1 (2.4) 7.8 (1.5) 7.9 (1.8) 2.1 (1.3)Arousal 5.6 (2.5) 3.0 (2.5) 6.1 (3.1) 4.4 (2.8) 2.4 (2.3) 5.9 (2.9)

Notes: Data are M (SD).

Please cite this article as: Patron, E., et al., Depression is associated with incardiac surgery, Auton. Neurosci. (2015), http://dx.doi.org/10.1016/j.autn

other contrasts (ps N .21, Cohen's ds b .28), was noted. In patientswithout depression, no difference among conditions was found(ps N .09, Cohen's ds b 0.45). The main effect for Condition (F[3,75] = 2.04, p = .13, ε = .85, η2p = .08) was not significant. Similarly,theGroup× Condition ANCOVA on lnHF yielded amarginally significantGroup × Condition interaction effect (F[3, 75] = 2.52, p = .08, ε = .82,η2p = .09), showing the same trend observed for HF componentexpressed in n.u. No significant main effect for Group (F[1, 25] = 1.78,p = .19, η2p = .07) or Condition (F[3, 75] = 1.47, p = .24, ε = .82,η2p = .06) was noted. Finally, the Group × Condition ANCOVA on lnLFdid not show significant effect for Group (F[1, 25] = 0.00, p = .99,η2p = .00), Condition (F[3, 75] = 1.59, p = .20, ε = .87, η2p = .06), orGroup × Condition interaction (F[3, 75] = 0.72, p = .52, ε = .87,η2p = .03). The results of IBIs and HRV analyses are shown in Fig. 1.

Correlation analyses revealed significant associations between CES-D scores and residualized changes in IBIs (r = − .38; p b .05) and HFn.u. (r = − .49; p b .01) during the imaging of the unpleasant scriptrelative to rest, as shown in Fig. 2.1 Specifically, greater CES-D scoreswere associated with greater reductions in IBIs and HF n.u. duringthe imaging of the unpleasant script. By contrast, CES-D scores wereunrelated to residualized changes in IBIs and HF n.u. during the imagingof the neutral and the pleasant scripts relative to rest (ps N .08). Nosignificant correlations between CES-D scores and changes in lnLF andlnHF were noted (ps N .14).

4. Discussion

The present study investigated the influence of depression on IBIsand HRV activity during an emotional imagery task in patients aftercardiac surgery. Consistent with the intended manipulation, self-reported ratings of valence were lower for the unpleasant than thepleasant and the neutral scripts in both groups, with a large effect sizeof the mean difference between unpleasant vs. pleasant (Cohen'sds ≥ 1.68) and unpleasant vs. neutral (Cohen's ds ≥ 1.06) scripts.Similarly, although marginally significantly, self-reported ratings ofarousal were greater for the pleasant and the unpleasant than theneutral scripts, with a large effect size of the mean difference betweenpleasant vs. neutral (Cohen's d = 0.91) and unpleasant vs. neutral(Cohen's d = 1.24) scripts. As expected, employing an emotionalimagery task, including pleasant, neutral and unpleasant scripts, waseffective in eliciting positive (i.e., higher valence and higher arousalratings in pleasant than neutral script) and negative (i.e., lower valenceand higher arousal ratings in unpleasant than neutral script) emotions(Bradley, 2000).

At the physiological level, we found that postoperative depressionwas directly associated with increased vagal withdrawal during theunpleasant emotional imagery. Indeed, our findings showed thatpatients with postoperative depression had significantly lower HF n.u.,reflecting greater vagal withdrawal, during the imaging of the unpleas-ant script compared to nondepressed patients after surgery. Consistentwith this finding, CES-D scores were inversely associated withresidualized changes in HF n.u. from rest to the imaging of the unpleas-ant script, but not the pleasant or the neutral one. Moreover, the imag-ing of the unpleasant script elicited greater vagal withdrawal(i.e., reduced HF n.u.) in patients with depression compared to thepleasant script, whereas no differences among emotions in nonde-pressed patients were noted. Although not significantly, results obtain-ed for lnHF resemble the same pattern of vagal activity found for HF n.uduring emotional imagery as a function of depression. By contrast,depression was unrelated to IBIs and lnLF activity during the imagerytask. However, correlation analysis revealed that CES-D scores were

1 All significant correlations survived the adjustment for QMI score (p b .05). Moreover,the significant correlation between CES-D scores and residualized changes in HF n.u. fromrest to imaging of unpleasant script survived after the exclusion of two outliers (r= − .75,p b .001).

creased vagal withdrawal during unpleasant emotional imagery aftereu.2015.02.002

Page 5

Fig. 1. Mean (SE) IBIs, lnLF, lnHF, and HF n.u. activity as a function of emotional Condition (rest, pleasant, neutral and unpleasant scripts) in the group with depression and withoutdepression. Mixed ANCOVAs revealed a significant Group × Condition interaction effect (p b .04). Fisher's LSD post hoc test revealed that patients with depression had significantlylower HF n.u. during unpleasant script compared to patients without depression (p b .01, Cohen's d = 1.34). A significantly lower HF n.u. for unpleasant script compared to pleasantone (p b .03, Cohen's d = 0.55) in patients with depression was also noted. IBIs = inter-beat intervals in ms; lnLF = log of low frequency power (0.04 to 0.15 Hz) in ms2; lnHF = logof high frequency power (0.15 to 0.40 Hz) in ms2; HF n.u. = high frequency expressed in normalized units [HF / (total power-VLF) ∗ 100].

5E. Patron et al. / Autonomic Neuroscience: Basic and Clinical xxx (2015) xxx–xxx

significantly associatedwith greater reductions in IBIs (i.e., greater heartrate activity) from rest to the imaging of the unpleasant script.

These novel findings add to the literature on depression-relatedaltered cardiovascular reactivity by showing that a depression-increased vagal withdrawal relationship extends to patients aftercardiac surgery. In particular, these preliminary findings complementstudies that have documented increased vagal withdrawal duringnon-emotional stressors (e.g., forehead cold pressor or mental arith-metic task) in healthy individuals with depressed mood (Hughes andStoney, 2000) as well as in patients with cardiovascular diseases(Kibler and Ma, 2004). Most importantly, the present study suggeststhat increased vagal withdrawal to unpleasant emotions in post-surgical

Fig. 2. Correlation analyses between scores in the Center for Epidemiological Studies of Depressiand high frequency components expressed in normalized units (HF n.u.) (r = − .49; p b .01) f

Please cite this article as: Patron, E., et al., Depression is associated with incardiac surgery, Auton. Neurosci. (2015), http://dx.doi.org/10.1016/j.autn

patients may mediate the conferral of cardiac risk by depression(Blumenthal et al., 2003; Lespérance et al., 2000).

The main finding – the depression-related increased vagalwithdrawal for the unpleasant script – is in line with previous reportsshowing that depression is characterized by a mood-congruent bias to-ward negatively valenced affects or stimuli (Eizenman et al., 2003;Erickson et al., 2005; Mogg and Bradley, 1998; Murphy et al., 1999). Inturn, the mood-congruent bias toward negative information supportsthe maladaptive patterns of information that trigger and sustaindepressed mood (Dalgleish and Watts, 1990; Mathews and MacLeod,1994; Mogg and Bradley, 1998). Consistent with this finding, it hasbeen proposed the negative potentiation view, which postulates that

on (CES-D) scale and residualized changes in inter-beat intervals (IBIs) (r=− .38; p b .05)rom rest to the imaging of unpleasant script.

creased vagal withdrawal during unpleasant emotional imagery aftereu.2015.02.002

Page 6

6 E. Patron et al. / Autonomic Neuroscience: Basic and Clinical xxx (2015) xxx–xxx

emotional and physiological reactivity to negative emotional stimuliis potentiated by the pervasive negative mood states that are fre-quent in patients with depression (Golin et al., 1977; Lewinsohnet al., 1973).

It should be noted, however, that our results are at odds with agrowing body of research reporting that depression is associated withattenuated cardiac increase and reduced vagal withdrawal duringpsychophysiological tasks, such as the speech task or the mentalarithmetic task, in medically healthy individuals (e.g., Carroll et al.,2007; Schwerdtfeger and Rosenkaimer, 2011). Given the evidenceshowing that vagal suppression is an important substrate for adaptationand flexible behavior in medically healthy individuals (Porges, 1995,1997), the lack of vagal withdrawal in response to stressors has beenproposed to reflect an impaired ability to flexibly respond to environ-mental stimuli in depressed individuals without medical illness(Rottenberg et al., 2007). By contrast, in patients with cardiovasculardisease, depression has been associated with increased rather thanattenuated cardiovascular reactivity and vagal withdrawal in responseto psychophysiological tasks (Kibler and Ma, 2004). Given that cardiacvagal control is dramatically decreased in patients with cardiovasculardisease or after cardiac surgery compared to healthy individuals(Soares et al., 2005), it can be argued that depression-related increasedvagal withdrawal in response to negative emotions is more likely torepresent a dysfunctional rather than a functional pattern of autonomicmodulation. Taken together, our results suggest that depression-relatedvagal abnormalities may differ as a function of the population in whichthey are studied.

In addition, our findings diverge from those by York et al. (2007),who reported that depressionwas associatedwith blunted cardiovascu-lar reactivity even in patients with CAD. It should be noted, however,that the discrepancy between our findings and those of York et al.(2007) may be accounted for by methodological differences. Specifical-ly, given that York et al. (2007) analyzed the correlation betweendepression and the change in cardiovascular parameters (i.e., heartrate, systolic and diastolic blood pressure) from baseline (i.e., pre-stress) to recovery period (i.e., post-stress), they examined cardiovascu-lar recovery after a stressor. Conversely, given thatwewere interested inexamining the influence of depression on cardiovagal activity duringemotional activation, heart rate and HRV parameters was analyzedduring different emotional conditions.

The relative contribution of the sympathetic and parasympatheticnervous systems to depression-related altered cardiac reactivity as arisk factor for cardiovascular diseases is still debated (Carney et al.,2005). Nonetheless, the current findings suggest that depression ismore likely to be associated with reduced cardiac vagal modulationrather than excessive sympathetic activity on heart. Indeed, a selectiveassociation was found between depression and HRV parameters thathave been shown to be largely influenced by cardiac vagal control(i.e., lnHF and HF n.u.) (Berntson et al., 1997; Taylor et al., 1998). Bycontrast, LF components, which are the mirror of both sympatheticand vagal influences on cardiac activity (Task Force, 1996), were unre-lated to depression after cardiac surgery. A possible explanation forthis finding is that cardiac sympathetic activity was at least partiallyattenuated by beta-blockers, which also may explain the null findingconcerning IBIs reactivity during the imagery task. However, it is impor-tant to note that CES-D scores were selectively associated with greaterreductions in IBIs from rest to the imaging of unpleasant script, thussuggesting that depression does increase heart rate activity in negativecontext.

From a clinical point of view, our study also suggests that biobehav-ioral interventions, such as HRV biofeedback, may offer an importantrehabilitation option for improving cardiac vagal tone at rest as well ascardiac vagal modulation during stressors in post-surgical patients(Nolan et al., 2005; Patron et al., 2013b). In particular, Nolan et al.(2005) have shown the efficacy of HRV-biofeedback intervention forincreasing vagal recovery from stress reactivity protocol in patients

Please cite this article as: Patron, E., et al., Depression is associated with incardiac surgery, Auton. Neurosci. (2015), http://dx.doi.org/10.1016/j.autn

with CAD. It is worth noting that the same authors also observed thatincreased cardiac vagal activity during stress recovery was inverselyassociatedwith depressive symptoms in those patientswho underwentbiofeedback of HRV compared to controls (i.e., without biofeedback).More recently, Patron et al. (2013b) have extended the effectivenessof HRV-biofeedback for increasing cardiac vagal control at rest as wellas for reducing depressive symptoms to the post-surgical period.Based on these results, future studies are warranted to test whethereffects of HRV-biofeedback also generalize from resting states to (emo-tional) stressful periods in patients after cardiac surgery.

The current findings should be interpreted in light of a number ofpossible methodological issues. First, this study used a relatively smallsample size; therefore, the present results need to be replicated tofully understand the relationship between postoperative depressionand cardiac reactivity during (emotional) stressors in patients aftercardiac surgery. However, the effect sizes for HF n.u. were medium tolarge, with the largest effect sizes tending to occur between groupsduring the imaging of the unpleasant script (Cohen's d = 1.34) aswell as between the pleasant and the unpleasant scripts within individ-uals with depression (Cohen's d = 0.55). Second, the current study didnot take into account respiration rate, even though the HRV may beconfounded and exaggerated by respiration without reflecting theinfluence of vagal tone on the HR (Grossman and Taylor, 2007). Howev-er, there is large evidence that even uncontrolled HRV recordings canbe an independent predictor of adverse outcome in medically ill popu-lations, especially in patients with cardiovascular diseases (Biggeret al., 1993; Kleiger et al., 1987; Rich et al., 1988). Moreover, giventhat imagery task did not require patients to speak or read, it is likelythat its influence on respiratory rate was minimal. Third, HRV may beunstable, especially in the early postoperative period. While HRV ismaximally reduced on about the sixth day after surgery, it graduallyrecovers, returning to preoperative values approximately two monthsafter cardiac surgery (Demirel et al., 2002; Soares et al., 2005). However,it is worth noting that the depression-related reduced HRVwas evidenteven in the early postoperative period (Patron et al., 2012, 2014), thussuggesting that depression may reduce HRV in addition to or indepen-dently of surgical procedures. In addition, depression was evaluatedby means of CES-D questionnaire but not clinically evaluated in thepatients enrolled in the study. Although CES-D has shown good psycho-metric properties and has been validated in a large community popula-tion, it cannot replace a psychiatric evaluation using structured criteriadefined in the fifth edition of the Diagnostic and Statistical Manual ofMental Disorders (DSM-5) (American Psychiatric Association, 2013).Finally, CES-D contains items that may reflect somatic symptomsassociated with cardiac disease and/or surgery rather than depressionitself, such as poor appetite, fatigue, insomnia, and inattentiveness.However, given that the groupwith depression andwithout depressiondid not differ in terms of surgical procedures and biomedical risk factors(e.g., hypertension, previous myocardial infarction), it is unlikely thatthe CES-D scores were inflated by somatic symptoms associated withcardiac disease and/or surgery differently in the group with vs. withoutdepression. Of note, the CES-D questionnaire has also shown adequatevalidity and reliability for use in patients with medical comorbidities(e.g., Blumenthal et al., 2003). Nonetheless, future research iswarrantedto replicate and extend these findings by selectively studying patientswith diagnosis of major depressive disorder.

5. Conclusions

The present study is, to our knowledge, the first to examinewhether and how cardiac activity to emotional stimuli is modulatedby depression in patients after cardiac surgery. Depressionwas associat-ed with greater vagal withdrawal during the imaging of the unpleasantscript. This finding therefore suggests that depression-related vagalwithdrawal during emotional stressors may be implicated as oneplausible pathophysiological mechanism linking depression and cardiac

creased vagal withdrawal during unpleasant emotional imagery aftereu.2015.02.002

Page 7

7E. Patron et al. / Autonomic Neuroscience: Basic and Clinical xxx (2015) xxx–xxx

risk in patients who underwent cardiac surgery. Finally, our findingssuggest the importance to include novel biobehavioral interventions,in order to add to the efficacy of cardiovascular risk reduction programsand/or cardiac rehabilitation.

Conflicts of interest

All authors declare that they have no conflicts of interest in thepresent study.

Acknowledgments

Funds for this study were provided by Motta di Livenza Hospital,Treviso, for the PhD student grant for Ms. Patron (Disturbi affettivi ecognitivi nelle patologie neuro-cardiovascolari), and by the EuropeanSocial Fund (ref. 2105/101/1/722/2009) for the PhD student grant forDr. Messerotti Benvenuti.

Appendix A

A.1. Pleasant script (narrative 8500)

A.1.1. English originalThe registered letter says that “You have just won ten million

dollars!” It's amazing — You bought the winning ticket in the lottery.You cry, scream, jump with joy!

A.1.2. Italian translation and adaptationAlla televisione stanno annunciando il biglietto che ha vinto il primo

premio della Lotteria, il numero corrisponde al biglietto che hocomprato tempo fa, non ci posso credere. Controllo più volte, ma èproprio quello vincente. E' incredibile, entusiasmante, ho davverovinto 5 milioni di euro. Mi sento così eccitato ed elettrizzato che miviene voglia di gridare a squarciagola, rido, salto e ballo per la gioia.

A.2. Neutral script (narrative 2610)

A.2.1. English originalYou are sitting at the kitchen table with yesterday's newspaper in

front of you. You push back the chair when you hear the coffee makerslow to a stop.

A.2.2. Italian translation and adaptationSono seduto rilassato sulla poltrona in soggiorno e sto leggendo il

giornale di ieri. Sento rumori familiari che provengono dalla cucina.Guardo fuori dalla finestra, è una giornata d'autunno, foglie rosse emarroni cadono dagli alberi e per strada passano alcune macchine.Dalla cucina sento arrivare il profumo del caffè appena fatto, mi sentotranquillo, con calma mi alzo per andare in cucina a bere il caffè.

A.3. Unpleasant script (narrative 6020)

A.3.1. English originalWithout thinking, you stepped off the curb into traffic. Breaks

screech. You look up, frozen, heart jumping in your chest. A truck isskidding, hurtling towards you.

A.3.2. Italian translation and adaptationSento uno stridore di freni, a pochi metri da me il mio amico giace a

terra investito da un'auto in corsa. Ha un profondo squarcio sulla testa, ilsangue esce a fiotti. Mi precipito verso di lui per aiutarlo, ma sono cosìspaventato che non riesco a fare nulla. Sto tremando, mi sento il cuorein gola, mi manca il respiro e la voce per chiamare aiuto.

Please cite this article as: Patron, E., et al., Depression is associated with incardiac surgery, Auton. Neurosci. (2015), http://dx.doi.org/10.1016/j.autn

References

American Psychiatric Association, 2013. Diagnostic and Statistical Manual of MentalDisorders. 5th ed. (Washington, DC).

Appelhans, B., Luecken, L., 2006. Heart rate variability as an index of regulated emotionalresponding. Rev. Gen. Psychol. 10, 229–240.

Barefoot, J.C., Helms, M.J., Mark, D.B., Blumenthal, J.A., Califf, R.M., Haney, T.L., O'Connor,C.M., Siegler, I.C., Williams, R.B., 1996. Depression and long-term mortality risk inpatients with coronary artery disease. Am. J. Cardiol. 78, 613–617.

Berntson, G.G., Cacioppo, J.T., Binkley, P.F., Uchino, B.N., Quigley, K.S., Fieldstone, A.,1994. Autonomic cardiac control. III. Psychological stress and cardiac responsein autonomic space as revealed by pharmacological blockades. Psychophysiology31, 599–608.

Berntson, G.G., Bigger Jr., J.T., Eckberg, D.L., Grossman, P., Kaufmann, P.G., Malik, M.,Nagaraja, H.N., Porges, S.W., Saul, J.P., Stone, P.H., van der Molen, M.W., 1997. Heartrate variability: origins, methods, and interpretive caveats. Psychophysiology 34,623–648.

Bigger Jr., J.T., Fleiss, J.L., Rolnitzky, L.M., Steinman, R.C., 1993. Frequency domainmeasuresof heart period variability to assess risk late after myocardial infarction. J. Am. Coll.Cardiol. 21, 729–736.

Blumenthal, J.A., Lett, H.S., Babyak, M.A., White, W., Smith, P.K., Mark, D.B., Jones, R.,Mathew, J.P., Newman, M.F., 2003. Depression as a risk factor for mortality aftercoronary artery bypass surgery. Lancet 362, 604–609.

Bradley, M.M., 2000. Emotion and motivation. In: Cacioppo, J.T., Tassinary, L.G., Berntson,G. (Eds.), Handbook of Psychophysiology. Cambridge University Press, New York,pp. 602–642.

Bradley, M.M., Lang, P.J., 1994. Measuring emotion: the Self-Assessment Manikin and theSemantic Differential. J. Behav. Ther. Exp. Psychiatry 25, 49–59.

Bradley, M.M., Lang, P.J., 2007. Affective Norms for English Text (ANET): Affective Ratingsof Text and Instruction Manual. (Technical Report No. D-1). University of Florida,Gainesville, FL.

Carney, R.M., Rich, M.W., Freedland, K.E., Saini, J., teVelde, A., Simeone, C., Clark, K., 1988.Major depressive disorder predicts cardiac events in patients with coronary arterydisease. Psychosom. Med. 50, 627–633.

Carney, R.M., Blumenthal, J.A., Stein, P.K., Watkins, L., Catellier, D., Berkman, L.F.,Czajkowski, S.M., O'Connor, C., Stone, P.H., Freedland, K.E., 2001. Depression, heartrate variability, and acute myocardial infarction. Circulation 104, 2024–2028.

Carney, R.M., Freedland, K.E., Miller, G.E., Jaffe, A.S., 2002. Depression as a risk factor forcardiac mortality and morbidity: a review of potential mechanisms. J. Psychosom.Res. 53, 897–902.

Carney, R.M., Freedland, K.E., Veith, R.C., 2005. Depression, the autonomic nervoussystem, and coronary heart disease. Psychosom. Med. 67, S29–S33.

Carroll, D., Phillips, A.C., Hunt, K., Der, G., 2007. Symptoms of depression and cardiovascu-lar reactions to acute psychological stress: evidence from a population study. Biol.Psychol. 75, 68–74.

Cohen, J., 1988. Statistical Power Analysis for the Behavioral Sciences. Lawrence Erlbaum,Hillsdale, NJ.

Cuthbert, B.N., Vrana, S.R., Bradley, M., 1991. Imagery: function and physiology. In: Ackles,P.K., Jennings, J.H., Coles, M.G.H. (Eds.), Advances in Psychophysiology. Jessica Kings-ley Publishers, London, pp. 1–42.

Dalgleish, T., Watts, F.N., 1990. Biases of attention andmemory in disorders of anxiety anddepression. Clin. Psychol. Rev. 10, 589–604.

Demirel, S., Akkaya, V., Oflaz, H., Tükek, T., Erk, O., 2002. Heart rate variability aftercoronary artery bypass graft surgery: a prospective 3-year follow-up study. Ann.Noninvasive Electrocardiol. 7, 247–250.

Eizenman, M., Yu, L.H., Grupp, L., Eizenman, E., Ellenbogen, M., Gemar, M., Levitan, R.D.,2003. A naturalistic visual scanning approach to assess selective attention in majordepressive disorder. Psychiatry Res. 118, 117–128.

Erickson, K., Drevets, W.C., Clark, L., Cannon, D.M., Bain, E.E., Zarate Jr., C.A., Charney, D.S.,Sahakian, B.J., 2005. Mood-congruent bias in affective go/no-go performance ofunmedicated patients with major depressive disorder. Am. J. Psychiatry 162,2171–2173.

Fava, G.A., 1982. Versione italiana del CES-D per la valutazione degli stati depressivi. In:Canestrari, R. (Ed.), Nuovi metodi in psicometria. Organizzazioni Speciali, Firenze,pp. 87–93.

Golin, S., Hartman, S.A., Klatt, E.N., Munz, K., Wolfgang, G.L., 1977. Effects of self-esteemmanipulation on arousal and reactions to sadmodels in depressed and nondepressedcollege students. J. Abnorm. Psychol. 86, 435–439.

Gollnisch, G., Averill, J.R., 1993. Emotional imagery: strategies and correlates. Cogn. Emot.7, 407–429.

Grossman, P., Taylor, E.W., 2007. Toward understanding respiratory sinus arrhythmia:relations to cardiac vagal tone, evolution and biobehavioral functions. Biol. Psychol.74, 263–285.

Hughes, J.W., Stoney, C.M., 2000. Depressed mood is related to high-frequency heart ratevariability during stressors. Psychosom. Med. 62, 796–803.

Kamath, M.V., Ghista, D.N., Fallen, E.L., Fitchett, D., Miller, D., McKelvie, R., 1987. Heart ratevariability power spectrogram as a potential noninvasive signature of cardiac regula-tory system response, mechanisms, and disorders. Heart Vessels 3, 33–41.

Kemp, A.H., Quintana, D.S., Gray, M.A., Felmingham, K.L., Brown, K., Gatt, J.M., 2010.Impact of depression and antidepressant treatment on heart rate variability: a reviewand meta-analysis. Biol. Psychiatry 67, 1067–1074.

Kibler, J.L., Ma, M., 2004. Depressive symptoms and cardiovascular reactivity to laboratorybehavioral stress. Int. J. Behav. Med. 11, 81–87.

Kleiger, R.E., Miller, J.P., Bigger Jr., J.T., Moss, A.J., 1987. Decreased heart rate variability andits association with increased mortality after acute myocardial infarction. Am.J. Cardiol. 59, 256–262.

creased vagal withdrawal during unpleasant emotional imagery aftereu.2015.02.002

Page 8

8 E. Patron et al. / Autonomic Neuroscience: Basic and Clinical xxx (2015) xxx–xxx

Lang, P.J., 1979. A bio-informational theory of emotional imagery. Psychophysiology 16,495–512.

Lang, P.J., Bradley, M.M., Cuthbert, B.N., 1990. Emotion, attention, and the startle reflex.Psychol. Rev. 97, 377–395.

Lespérance, F., Frasure-Smith, N., Juneau, M., Théroux, P., 2000. Depression and 1-yearprognosis in unstable angina. Arch. Intern. Med. 160, 1354–1360.

Lewinsohn, P.M., Lobitz, W.C., Wilson, S., 1973. “Sensitivity” of depressed individuals toaversive stimuli. J. Abnorm. Psychol. 81, 259–263.

Malliani, A., Lombardi, F., Pagani, M., 1994. Power spectrum analysis of heart rate variabil-ity: a tool to explore neural regulatory mechanisms. Br. Heart J. 71, 1–2.

Mathews, A., MacLeod, C., 1994. Cognitive approaches to emotion and emotionaldisorders. Annu. Rev. Psychol. 45, 25–50.

Messerotti Benvenuti, S., Zanatta, P., Longo, C., Mazzarolo, A.P., Palomba, D., 2012a.Preoperative cerebral hypoperfusion in the left, not in the right, hemisphere isassociated with cognitive decline after cardiac surgery. Psychosom. Med. 74, 73–80.

Messerotti Benvenuti, S., Zanatta, P., Valfrè, C., Polesel, E., Palomba, D., 2012b. Preliminaryevidence for reduced preoperative cerebral blood flow velocity as a risk factor forcognitive decline three months after cardiac surgery: an extension study. Perfusion27, 486–492.

Messerotti Benvenuti, S., Patron, E., Zanatta, P., Polesel, E., Bonfà, C., Palomba, D., 2013.Change in behavioral functional capacity is associated with preexisting cognitivefunction rather than with cognitive decline in patients 1 year after cardiac surgery.Gen. Hosp. Psychiatry 35, 117–121.

Mogg, K., Bradley, B.P., 1998. A cognitive-motivational analysis of anxiety. Behav. Res.Ther. 36, 809–848.

Murphy, F.C., Sahakian, B.J., Rubinsztein, J.S., Michael, A., Rogers, R.D., Robbins, T.W.,Paykel, E.S., 1999. Emotional bias and inhibitory control processes in mania anddepression. Psychol. Med. 29, 1307–1321.

Musselman, D.L., Evans, D.L., Nemeroff, C.B., 1998. The relationship of depression tocardiovascular disease: epidemiology, biology, and treatment. Arch. Gen. Psychiatry55, 580–592.

Nolan, R.P., Kamath, M.V., Floras, J.S., Stanley, J., Pang, C., Picton, P., Young, Q.R., 2005.Heart rate variability biofeedback as a behavioral neurocardiac intervention toenhance vagal heart rate control. Am. Heart J. 149, 1137.

Patron, E., Messerotti Benvenuti, S., Favretto, G., Valfrè, C., Bonfà, C., Gasparotto, R.,Palomba, D., 2012. Association between depression and heart rate variability inpatients after cardiac surgery: a pilot study. J. Psychosom. Res. 73, 42–46.

Patron, E., Messerotti Benvenuti, S., Zanatta, P., Polesel, E., Palomba, D., 2013a. Preexistingdepressive symptoms are associated with long-term cognitive decline in patientsafter cardiac surgery. Gen. Hosp. Psychiatry 35, 472–479.

Patron, E., Messerotti Benvenuti, S., Favretto, G., Valfrè, C., Bonfà, C., Gasparotto, R.,Palomba, D., 2013b. Biofeedback assisted control of respiratory sinus arrhythmia asa biobehavioral intervention for depressive symptoms in patients after cardiacsurgery: a preliminary study. Appl. Psychophysiol. Biofeedback 38, 1–9.

Patron, E., Messerotti Benvenuti, S., Favretto, G., Gasparotto, R., Palomba, D., 2014.Depression and reduced heart rate variability after cardiac surgery: the mediatingrole of emotion regulation. Auton. Neurosci. 180, 53–58.

Podrid, P.J., Fuchs, T., Candinas, R., 1990. Role of the sympathetic nervous system in thegenesis of ventricular arrhythmia. Circulation 82, I103–I113.

Please cite this article as: Patron, E., et al., Depression is associated with incardiac surgery, Auton. Neurosci. (2015), http://dx.doi.org/10.1016/j.autn

Porges, S.W., 1995. Orienting in a defensive world: mammalian modifications of ourevolutionary heritage. A polyvagal theory. Psychophysiology 32, 301–318.

Porges, S.W., 1997. Emotion: an evolutionary by-product of the neural regulation of theautonomic nervous system. Ann. N. Y. Acad. Sci. 807, 62–77.

Radloff, L.S., 1977. The CES-D Scale: a self-report depression scale for research in thegeneral population. Appl. Psychol. Meas. 1, 385–401.

Rich, M.W., Saini, J.S., Kleiger, R.E., Carney, R.M., teVelde, A., Freedland, K.E., 1988.Correlation of heart rate variability with clinical and angiographic variables and latemortality after coronary angiography. Am. J. Cardiol. 62, 714–717.

Rottenberg, J., Clift, A., Bolden, S., Salomon, K., 2007. RSA fluctuation in major depressivedisorder. Psychophysiology 44, 450–458.

Salomon, K., Clift, A., Karlsdóttir, M., Rottenberg, J., 2009. Major depressive disorder isassociated with attenuated cardiovascular reactivity and impaired recovery amongthose free of cardiovascular disease. Health Psychol. 28, 157–165.

Schwartz, P.J., Vanoli, E., 1981. Cardiac arrhythmias elicited by interaction between acutemyocardial ischemia and sympathetic hyperactivity: a new experimental model forthe study of antiarrhythmic drugs. J. Cardiovasc. Pharmacol. 3, 1251–1259.

Schwerdtfeger, A., Rosenkaimer, A.K., 2011. Depressive symptoms and attenuatedphysiological reactivity to laboratory stressors. Biol. Psychol. 87, 430–438.

Sheehan, P.W., 1967. A shortened form of Betts' questionnaire upon mental imagery.J. Clin. Psychol. 23, 386–389.

Sheffield, D., Krittayaphong, R., Cascio, W.E., Light, K.C., Golden, R.N., Finkel, J.B., Glekas, G.,Koch, G.G., Sheps, D.S., 1998. Heart rate variability at rest and during mental stress inpatients with coronary artery disease: differences in patients with high and lowdepression scores. Int. J. Behav. Med. 5, 31–47.

Soares, P.P., Moreno, A.M., Cravo, S.L., Nóbrega, A.C., 2005. Coronary artery bypass surgeryand longitudinal evaluation of the autonomic cardiovascular function. Crit. Care 9,R124–R131.

Task Force of the European Society of Cardiology and the North American Society forPacing and Electrophysiology, 1996. Heart rate variability: standards of measure-ment, physiological interpretation, and clinical use. Circulation 93, 1043–1065.

Taylor, J.A., Carr, D.L., Myers, C.W., Eckberg, D.L., 1998. Mechanisms underlying very-low-frequency RR-interval oscillations in humans. Circulation 98, 547–555.

vanOyen Witvliet, C., Vrana, S.R., 1995. Psychophysiological responses as indices ofaffective dimensions. Psychophysiology 32, 436–443.

Vrana, S.R., Spence, E.L., Lang, P.J., 1988. The startle probe response: a new measure ofemotion? J. Abnorm. Psychol. 97, 487–491.

Williams, S.E., Cumming, J., 2012. Challenge vs. threat: investigating the effect of usingimagery to manipulate stress appraisal of a dart throwing task. Sport Exerc. Psychol.Rev. 8, 4–21.

Williams, S.E., Cumming, J., Balanos, G.M., 2010. The use of imagery to manipulatechallenge and threat appraisal states in athletes. J. Sport Exerc. Psychol. 32, 339–358.

Wulsin, L.R., Singal, B.M., 2003. Do depressive symptoms increase the risk for theonset of coronary disease? A systematic quantitative review. Psychosom. Med.65, 201–210.

York, K.M., Hassan, M., Li, Q., Li, H., Fillingim, R.B., Sheps, D.S., 2007. Coronary arterydisease and depression: patients with more depressive symptoms have lowercardiovascular reactivity during laboratory-induced mental stress. Psychosom. Med.69, 521–528.

creased vagal withdrawal during unpleasant emotional imagery aftereu.2015.02.002