Heart rate variability biofeedback as a behavioral neurocardiac intervention to enhance vagal heart rate control

Prevention and Rehabilitation

Heart rate variability biofeedback as a behavioral neurocardiac intervention to enhance vagal heart rate control Robert P. Nolan, PhD,a Markad V. Kamath, PhD,b John S. Floras, MD, DPhil,c Jill Stanley, MA,d Clement Pang, MSc,b Peter Picton, MASc,c and Quincy R. Young, PhDe Toronto, Ontario, Hamilton, Ontario, and Vancouver, British Columbia, Canada

Background Patients with coronary heart disease (CHD) who experience depressed mood or psychological stress exhibit decreased vagal control of heart rate (HR), as assessed by spectral analysis of HR variability (HRV). Myocardial infarction and sudden cardiac death are independently associated with depression and stress, as well as impaired vagal HR control. This study examined whether a behavioral neurocardiac intervention to reduce stress or depression can augment cardiovagal modulation in CHD patients. We hypothesized that (1) cognitive-behavioral training with HRV biofeedback would augment vagal recovery from acute stress, and (2) vagal regulation of HR would be inversely associated with stress and depression after treatment. Methods

This randomized controlled trial enrolled 46 CHD patients from 3 clinics of CHD risk reduction in Toronto and Vancouver, Canada. Subjects were randomized to five 1.5-hour sessions of HRV biofeedback or an active control condition. Outcome was assessed by absolute and normalized high-frequency spectral components (0.15 - 0.50 Hz) of HRV, and by the Perceived Stress Scale and Centre for Epidemiologic Studies in Depression scale.

Results

Both groups reduced symptoms on the Perceived Stress Scale ( P = .001) and Centre for Epidemiologic Studies in Depression scale ( P = .004). Hierarchical linear regression determined that improved psychological adjustment was significantly associated with the high-frequency index of vagal HR modulation only in the HRV biofeedback group. Adjusted R 2 was as follows: HRV biofeedback group, 0.86 for stress ( P = .02) and 0.81 for depression ( P = .03); versus the active control group, 0.04 ( P = .57) and 0.13 ( P = .95), respectively.

Conclusion A novel behavioral neurocardiac intervention, HRV biofeedback, can augment vagal HR regulation while facilitating psychological adjustment to CHD. (Am Heart J 2005;149:1137.e1- 1137.e7.) Depressed mood and psychological stress are independently associated with primary and secondary coronary events.1,2 The weight of evidence also indicates that these suspected risk factors are associated with

From the aBehavioural Cardiology Research Unit, University Health Network and Faculty of Medicine, University of Toronto, Toronto, Canada, bFaculty of Medicine, Health Sciences Centre, McMaster University, Hamilton, Canada, cDepartment of Medicine, and the University of Toronto, University Health Network and Mount Sinai Hospital, Toronto, Canada, dDepartment of Psychology, York University, Toronto, Canada, and e Cardiac Program, St Paul’s Hospital, Vancouver, Canada. This research was supported by grant no. NA4126 from the Heart and Stroke Foundation of Canada (RPN) and by funding from the deGroote Foundation and Natural Sciences and Engineering Research Council of Canada (MVK). JSF holds the Canada Research Chair in Integrative Cardiovascular Biology and is a Career Investigator of the Heart and Stroke Foundation of Ontario. Submitted July 20, 2004; accepted March 11, 2005. Reprint requests: Robert Nolan, PhD, CPsych, Behavioural Cardiology Research Unit, NU 6N-618, University Health Network, 585 University Avenue, Toronto, Ontario, Canada M5G 2N2. E-mail: [email protected] 0002-8703/$ - see front matter n 2005, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2005.03.015

impaired neurocardiac regulation. Markers of vagal heart rate (HR) control are decreased among cardiac and noncardiac samples who are depressed3 - 5 or who are exposed to psychological stress.6 - 8 There is also evidence that these factors act synergistically, as stressinduced vagal HR inhibition is greater among cardiac and noncardiac samples with more severe symptoms of depression.9,10 An association between vagal HR inhibition and depression is not observed consistently, as the inverse association between vagal HR control and depression was found only among patients with moderate to severe depression in 1 study,17 whereas another study reported only a statistical trend for this association.14 Greater sympathetic influence on HR has been observed among depressed patients without evidence of decreased vagal HR modulation.11 Finally, clinically depressed patients have demonstrated an inverse association between baroreflex sensitivity and symptoms of anxiety, but not depression.12 A major challenge confronting behavioral cardiology is to determine whether a reduction in symptoms of

American Heart Journal June 2005

1137.e2 Nolan et al

depression or stress can enhance vagal HR control. Impaired vagal HR regulation is an important predictor of stress-induced myocardial ischemia,7 as well as cardiac mortality after myocardial infarction.13 There are several challenges for antidepressant or stress-reducing pharmacotherapies. Benzodiazepines can exert a vagolytic effect.14 Tricyclic antidepressants are known to have anticholinergic effects.15 Norepinephrine reuptake inhibitors diminish HR variability (HRV) directly by increasing the local concentrations of norepinephrine at the sinoatrial node,16 while indirectly enhancing cardiovagal regulation by stabilizing psychological stress or depressed mood. Serotonin-norepinephrine reuptake inhibitors can also diminish HRV.17 On the other hand, selective serotonin reuptake inhibitors have shown inconsistent results in influencing global HRV or markers of cardiovagal regulation.18,19 By avoiding direct antagonistic effects on tonic or reflex vagal HR modulation, behavioral interventions may offer an important adjunctive or alternative treatment option for improving neurocardiac adjustment to coronary heart disease (CHD). To that end, we examined 2 hypotheses: (1) cognitive-behavioral training which used HRV biofeedback would increase vagal HR recovery after a stress reactivity protocol, and (2) enhanced vagal HR recovery would be associated with a reduction in psychological stress and depression.

Methods Study patients Ninety subjects were screened from the CHD risk reduction programs of 3 teaching hospitals in Toronto and Vancouver, Canada. Failure to confirm CHD status resulted in the loss of 9 subjects, whereas 33 subjects declined to participate because of scheduling for cardiac procedures, conflicting requirements with other research protocols, or unavailability. The remaining 46 patients were randomized at the participating sites. All subjects were diagnosed with CHD, as defined by myocardial infarction or positive diagnostic assessment noted in the medical record (eg, positive exercise stress test). The study protocol was approved by each hospital/university ethics committee, and subjects provided written informed consent for this investigation. Exclusion criteria were class III or IV CHF, second or third degree AV block, active unstable angina, atrial fibrillation, clinically significant arrhythmia, sick sinus syndrome, or valvular disease.

Intervention protocol Patients were randomized to receive 5 sessions of an active control intervention or HRV biofeedback. Sessions were scheduled within a 4 week interval, and each session lasted 1.5 hours. The active control condition was consistent with stress management protocols that are offered as usual care in programs of secondary prevention. The initial 30 minutes provided training in cognitive-behavioral skills to manage psychological stress, by means of countering negative cognitions, priming positive coping statements to manage stress

triggers, increasing pleasurable activities, and maintaining social supports. In the remaining hour, autogenic relaxation was taught to enhance psychological well-being and to modify symptoms of stress through the relaxation response. Autogenic relaxation was conducted by tape-recorded instruction and adapted from a standardized protocol.20 The HRV biofeedback intervention was designed to augment respiratory sinus arrhythmia and thereby counter vagal inhibition after acute stress.21-23 After the brief cognitive-behavioral training described above, patients were taught to counter acute stress with paced breathing at 6 breaths/min, guided by HRV biofeedback. The training goal was to produce a 0.10 -Hz peak in the online HRV power spectrum during stress recovery periods. The HRV display was computed from a sample of 140 to 200 R-R intervals and updated at successive 10-second periods throughout the recovery period following stressors. Only the low-frequency ( LF 0.04 - 0.15 Hz) and high-frequency ( HF: 0.15 - 0.40 Hz) spectral bandwidths were presented to subjects to highlight short-term neurocardiac regulation. Subjects were directed to use this countering skill to actively regulate their recovery after stress exposure, as opposed to relaxing passively during this period. The initial session trained patients to produce the 0.10 -Hz peak with entrained breathing in the absence of acute stress. Later sessions presented standardized stress tasks. Subjects were instructed to behaviorally counter each stressor during the recovery interval by producing the 0.10-Hz peak on the HRV biofeedback screen. The stress tasks used in treatment sessions included serial 7 subtraction, serial 8 addition, as well as the consonant trigrams test with addition by 2’s and subtraction by 3’s. Stress tasks were not repeated across treatment sessions or during the assessment before and after the intervention program to avoid practice effects.

Assessment protocol A stress reactivity protocol was administered before and after 5 treatment sessions. Subjects were assessed in a semireclined position. After adaptation to the laboratory, HRV was recorded over a 5-minute baseline. Patients were then exposed to 3 successive 5-minute stressors, which were randomized to control for order effects. The physical stressor involved standing from the semireclined position. The Paced Auditory Serial Addition Test (PASAT)24 served as a cognitive stressor. This test was administered by audiotaped instruction which progressively increased performance demands. The final stress event recall task required subjects to describe a personal stressor that they had rated as z7.5 on a Likert-type scale of 0 ( bnot at all stressful Q ) to 10 ( bextremely stressful Q ). A 5-minute recovery period followed each of the 3 stress tasks. During each recovery, subjects were instructed to breathe spontaneously and to not move, talk, or mentally pace their breathing rate. Similarly, after completing the 3 stress tasks, subjects were instructed to relax as deeply as possible, to remain silent, and to breathe spontaneously. The purpose of this stress-countering procedure was to assess the ability of patients to intentionally recover from stress while the confounding influence of slow or paced respiration was procedurally controlled. At no time during the assessments were subjects exposed to their HRV spectral profiles.

American Heart Journal Volume 149, Number 6

HRV analysis Electrocardiogram data were assessed from lead 2 of the electrocardiogram, using the LabVIEW software platform (National Instruments, Austin, Tex). Each 5-minute stress recovery period was analyzed offline in brief segments (c2.5 minutes). As previously reported,25,26 QRS complexes from the analog output of the electrocardiogram amplifier were processed with a microcomputer via a 12-bit analog-to-digital converter, with a sampling rate of 512 Hz and stored sequentially for spectral analysis. Extra or missing beats were replaced with R-R intervals calculated by linear interpolation from adjacent cycles. R-R intervals for each 2.5-minute segment were assessed with fast Fourier transform to calculate total spectral power, from which LF (0.04 - 0.15 Hz) and HF powers (0.15 - 0.50 Hz) were assessed in units of m2/Hz. HF power was also assessed in normalized units (nu), which were calculated as the proportion of HF relative to the sum of power within LF and HF bandwidths.

Measurements Symptoms of depressed mood and psychological stress were assessed by the Centre for Epidemiologic Studies in Depression (CES -D) scale27 and the Perceived Stress Scale (PSS).28

Statistical analysis Data were analyzed using SPSS version 10.1 (SPSS Inc, Chicago, Ill ). Results are reported as mean F SEM. Statistical analyses used log-transformed values of HF and HFnu to control for skewness. Statistical significance was interpreted from 2-tailed tests with P b .05. Differences between the 2 intervention groups with respect to baseline characteristics were examined by v 2 tests and 1-way analysis of variance (ANOVA). Repeated-measures ANOVAs were used to examine group differences in vagal recovery during the stress-countering procedure with self-guided relaxation, and during passive recovery from each of the 3 stressors in the assessment protocol (standing, PASAT, and stress event recall tasks). HF and HFnu were dependent variables. The Greenhouse-Geisser adjustment for asphericity was used. Post hoc comparisons were conducted with the Bonferroni correction for significance. The efficacy of these interventions in decreasing psychological stress and depression was assessed with a repeated-measures ANOVA. A hierarchical linear regression analysis examined whether logHF power was associated with stress and depression after intervention, while statistically controlling for the severity of stress and depression before the intervention.

Results Baseline characteristics Characteristics of the 46 subjects who were randomized to interventions at our 3 sites are presented in Table I. One subject was excluded because of incomplete psychometric assessment. Electrocardiogram data from the assessments before and after treatment were carefully reviewed. Three subjects were excluded because of technical problems (eg, arrhythmias). During the assessment session that

Nolan et al 1137.e3

Table I. Sociodemographic and medical characteristics of subjects at baseline

Characteristics Women/men Age, y BMI Education, y Income (bUS$50 000/ zUS$50 000) CES-D PSS Diabetes Hypertension Smokers MI only CABG/PTCA/ANG only MI and CABG/ PTCA/ANG Months since MI Months since CABG/PTCA/ANG h-Blockers, calcium antagonists, ACE inhibitors Antilipids Antiarrhythmics Anticoagulants Anxiolytics Antidepressants Change in medication during trial Dropout

HRV biofeedback (n = 27)

Active control (n = 19)

P

5/22 54.22 F 1.04 28.60 F 0.86 15.19 F 0.81 11/16

1/18 54.95 F 1.52 27.51 F 0.83 14.53 F 0.92 9/10

.20 .69 .38 .60 .66

17.99 F 2.02 18.96 F 1.06 5 8 10 6 10 11

15.43 F 2.27 17.83 F 1.25 3 7 7 1 8 9

.42 .50 .54 .41 .62 .12 .48 .44

22.00 F 13.47 2.20 F 0.66

59.96 F 33.57 9.77 F 4.95

.29 .10

22

16

.57

21 3 20 1 5 4

18 2 10 3 2 1

.12 .67 .20 .18 .38 .29

1/27

1/19

.80

Values are expressed as mean F SEM unless otherwise indicated. BMI, Body mass index; MI; myocardial infarction; ANG, angiogram; CABG, coronary artery bypass graft; PTCA, percutaneous transluminal coronary angioplasty; ACE, angiotensinconverting enzyme.

followed intervention, 8 subjects demonstrated an obvious spectral peak of b0.12 Hz, which is characteristic of breathing at a paced or slow rhythm. Exclusion of these subjects was necessary because of the confounding shift of spectral power from the HF to the LF bandwidth, and because unlike the intervention sessions, subjects were directed in the assessments to breathe spontaneously at a normal frequency (z0.15 Hz), even during self-guided relaxation. There were no significant differences between the intervention groups with respect to baseline characteristics or in attrition from their respective interventions.

Vagal recovery during stress countering Figure 1 shows values of logHF during baseline and the stress-countering procedure that followed exposure to the stress reactivity tasks, before and after intervention. Before intervention, logHF did not change significantly during baseline and the stress-countering procedure for the HRV biofeedback group (2.11 F 0.13 to 2.10 F 0.10,

American Heart Journal June 2005

1137.e4 Nolan et al

Figure 1

Table II. Hierarchical linear regression of psychological stress and depression on logHF after intervention

*

2.30

Std b

2.25

LogHF

2.20

*

2.15 2.10 2.05 2.00 BSLN

Countering

Before Intervention AC Group

BSLN

Countering

After Intervention HRV-BFB

BSLN, baseline; countering, self-guided relaxation immediately after exposure to stress reactivity tasks. Note that all logHF values are means F SEM averaged across the first and second 2.5-minute intervals of BSLN and countering procedure, given that repeated ANOVAs did not identify any significant interaction with the 2.5-minute intervals. *P = .050. AC, Active control; BFB, biofeedback.

respectively; F = 0.08, P = .78) and for the active = con trol group (2.05 F 0.08 to 2.12 F 0.09, respectively; F = 1.78, P = .20). After treatment, the HRV biofeedback group increased logHF from 2.07 F 0.11 at baseline to 2.19 F 0.10 during the stress-countering procedure ( F = 4.42, P b .050). The active control group did not demonstrate significant changes in logHF from baseline to the stress-countering procedure after intervention (2.12 F 0.11 to 2.12 F 0.11, respectively; F = 0.02, P = .90). There were no group differences for these tasks using repeated-measures ANOVA with logHFnu as the dependent variable.

Passive recovery from stress tasks Separate repeated-measures ANOVAs were conducted on logHF during each stress reactivity task and the recovery interval that immediately followed. Analysis of the physical stressor (standing) and recovery interval resulted in a higher order interaction: Active control versus HRV biofeedback groups
before versus after intervention
stress task versus recovery
first versus second 2.5-minute period ( F = 4.85, P = .03). Repeatedmeasures ANOVAs and Bonferroni-corrected post hoc comparisons indicated that before intervention, logHF did not change significantly across the first and second 2.5-minute periods of the physical stress task and recovery for either active controls ( F = 0.78, P = .39) or the HRV biofeedback group ( F = 1.41, P = .25). After the intervention, however, logHF power increased significantly for the HRV biofeedback group, from the first and second 2.5-minute periods of the

PSS HRV biofeedback group Model 1: before intervention PSS Model 2: before intervention PSS After intervention logHF-baseline After intervention logHF-SR first period After intervention logHF-SR second period Active control group Model 1: before intervention PSS Model 2: before intervention PSS After intervention logHF-baseline After intervention logHF-SR first period After intervention logHF-SR second period

.88T .67T .55y .12

Adjusted P for R 2 R2 change

.75 .86

b.001 .022

.10 .04

.131 .567

.68 .81

b.001 .033

.32 .13

.020 .946

.71z

.41 .32 .60 .27 .79

Depressed mood (CES-D) HRV biofeedback group Model 1: before intervention CES-D .84T Model 2: before intervention CES-D .70T After intervention logHF-baseline .71z After intervention logHF-SR .05 first period After intervention logHF-SR .56y second period Active control group Model 1: before intervention CES-D .61y Model 2: before intervention CES-D .58 After intervention logHF-baseline .09 After intervention logHF-SR .08 first period After intervention logHF-SR .01 second period

LogHF-baseline, Second 2.5-minute period; logHF-SR, stress-countering procedure with self-guided relaxation at first and second 2.5-minute periods. TP b .001. yP b .05. zP b .01.

physical stress task (1.84 F 0.10, 1.92 F 0.10, respectively) to the first and second 2.5-minute periods of recovery (2.27 F 0.10, 2.13 F 0.10, respectively; F = 7.34, P = .01). LogHF power did not change significantly for the active control group after intervention (F = 1.46, P = .25). Repeated-measures ANOVAs were conducted for LogHF during the cognitive stressor (PASAT) and stress event recall tasks and the respective recovery intervals. There were no significant main effects or interactions involving the intervention groups. Similarly, repeated ANOVAs were conducted for all 3 stress tasks and recovery periods with logHFnu as the dependent variable, and there were no significant differences involving the intervention groups.

American Heart Journal Volume 149, Number 6

Psychological adjustment and vagal recovery Symptoms of depression and psychological stress decreased significantly after intervention for all subjects: CES-D, 16.61 F 1.59 to 13.46 F 1.59 ( F = 8.48, P = .006); PSS, 18.32 F 0.84 to 15.96 F 0.83 ( F = 15.03, P b .001). The interaction between assessment intervals and intervention group was not significant in the case of depression ( F = 0.04, P = .84) and stress ( F = 0.16, P = .69). Hierarchical linear regression analyses were conducted separately for the HRV biofeedback and active control groups to assess the association between logHF and both depression and stress after intervention. As noted in Table II, logHF during the stress-countering procedure was inversely associated with the PSS for the HRV biofeedback group, after statistically controlling for PSS levels before intervention (adjusted R 2 = 0.86, P = .022). A similar inverse association was observed between logHF during stress countering and the CES-D, after controlling for CES-D scores before intervention (adjusted R 2 = 0.81, P = .033). In contrast, the active control group failed to demonstrate any association between logHF and either the PSS (adjusted R 2 = 0.04, P = .567) or CES-D (adjusted R 2 = 0.13, P = .946).

Discussion This study evaluated whether a behavioral neurocardiac intervention could enhance both vagal HR regulation and psychological adjustment to CHD. Previous findings indicate that vagal HR modulation is decreased among CHD patients with depression or during psychological stress.4,7,10 These psychological factors also independently predict primary and secondary coronary events.1,2 Given that markers of vagal inhibition independently predict cardiac death in CHD patients,13 it is plausible that impaired cardiovagal regulation is a mediating pathway between depression or psychological stress and coronary events. Two principal hypotheses were supported in this investigation. First, 5 sessions of HRV biofeedback with paced breathing and brief cognitive-behavioral training enhanced vagal HR regulation during self-guided recovery from a stress reactivity protocol and during passive recovery from a physical stressor. Second, patients receiving HRV biofeedback reduced symptoms of psychological stress and depression, and this improvement was associated with enhanced vagal HR modulation. The active control condition also reduced psychological stress and depression, but this outcome was not associated with vagal cardiac control. To our knowledge, this is the first investigation in which HRV biofeedback and paced breathing were combined with cognitive-behavioral training to determine if this augmented vagal HR modulation while reducing symptoms of psychological stress or depres-

Nolan et al 1137.e5

sion. Previous controlled trials that combined paced breathing with biofeedback reported an increase in HRV and BRS among subjects with CHD,29 or chronic obstructive pulmonary disease,30 or among healthy adults.31 HRV was also increased in an uncontrolled trial with CHD patients.32 A negative outcome was reported for a sample of asthmatic patients, although respiratory functioning improved with biofeedback.33 The critical pathways that mediate how biofeedback training augments HRV or BRS are not well understood. HRV biofeedback training with paced breathing likely reinforces peripheral HR modulation by the arterial baroreceptors,23 as well as by chemoreceptors, metabo receptors, and cardiopulmonary mechanoreceptors.22 Interestingly, Vaschillo et al34 reported that enhanced HRV after biofeedback training was independent of respiration in a single session study, and in a controlled trial, it was independent of both respiration and symptoms of relaxation.31 This observation is consistent with our results, in which vagal HR modulation after psychological stress increased after HRV biofeedback training, but not after training in autogenic relaxation. For now, there is insufficient data to determine if paced respiration or subjective relaxation is necessary or sufficient for the efficacy of HRV biofeedback. An adequate understanding of how HRV biofeedback influences vagal HR control requires a critical summary of how medullary neurons in the cardiorespiratory network are modulated by cortical frontal-limbic circuits within the central autonomic network—reviewed elsewhere.22,35 This is particularly important because biofeedback training requires subjects to maintain focused attention on a targeted physiological index (eg, a 0.10 -Hz peak in the HRV power spectrum), while regulating their emotional response. In essence, subjects are engaged in a goal-directed activity that is designed to increase their perceived efficacy and operant skill in regulating a physiological target through performance-based (bio)feedback that is provided online. Biofeedback may enhance vagal HR regulation by evoking focused concentration in combination with emotional selfcontrol. This cognitive-emotional response is associated with a neural circuit in which the prefrontal cortex (ventromedial and dorsolateral areas) and the anterior cingulate cortex (dorsal and ventral areas) play a prominent role, as the adaptive control of emotion and goal-directed behavior is initiated and maintained.35 Interestingly, these structures are also functionally linked to neurocardiac regulation through reciprocal interconnections with the insula, amygdala, lateral hypothalamus, parabrachial nucleus, and neural centers in the medulla involved in sympathetic and parasympathetic effector pathways to the heart.22,35 Preliminary evidence indicates that a meditation procedure can activate the prefrontal cortex while evoking increased HRV.36 This finding may have high

1137.e6 Nolan et al

relevance to HRV biofeedback research, and it suggests that the interface between cognitive-emotional functioning and neurocardiac regulation merits high priority in future biofeedback studies. HRV biofeedback with paced breathing may increase the efficacy of cognitive-behavioral treatments that are designed to facilitate adjustment to CHD. Conventional cognitive-behavioral programs enhance mood and affect by inhibiting negative self-talk and cognitive rumination on negative events, while increasing pleasurable social activities, subjective relaxation, and cognitive priming of positive emotions. The diversity of these strategies makes it difficult to know the neural pathways that might mediate their influence on cardiac functioning. This lack of specificity hinders the development of cognitive-behavioral therapy as a tool for primary or secondary prevention. The findings of our trial suggest that behavioral relaxation with brief cognitive-behavioral training is not sufficient to independently decrease sympathoexcitatory effects on vagal HR regulation, despite the fact that subjects may report decreased stress or depression. A limitation of this current study is the relatively small sample and the exclusion of some subjects because of the confounding influence of respiration (b0.12 Hz) during our assessments. In addition, the specific mechanical and reflex changes that mediated our treatment outcome were not examined. The cognitive-behavioral skills component of our intervention was consistent with stress reduction and depression management services that are offered in cardiac programs. We did not attempt to follow a single established protocol, however, which could make replication more difficult. Another limitation is that this study does not provide data to determine whether decreased depression and psychological stress, combined with increased vagal HR modulation, improved the long-term clinical outcome for our sample. A priority that emerges from this study is to determine if the positive effect of HRV biofeedback on vagal HR regulation and psychological functioning can be replicated with populations at risk for primary or secondary coronary events. This novel behavioral neurocardiac intervention may add to the efficacy of CHD risk reduction programs. We would like to thank Ms Rhoda Ortiz for assistance in preparing this manuscript. Dr Nolan has received honoraria from the Biofeedback Foundation of Europe.

References 1. Hemingway H, Marmot M. Evidence based cardiology: psychosocial factors in the aetiology and prognosis of coronary heart disease. Systematic review of prospective cohort studies. BMJ 1999;318:1460 - 7.

American Heart Journal June 2005

2. Rozanski A, Blumenthal JA, Kaplan J. Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy. Circulation 1999;99:2192 - 217. 3. Agelink MW, Boz C, Ullrich H, et al. Relationship between major depression and heart rate variability. Clinical consequences and implications for antidepressive treatment. Psychiatry Res 2002;113: 139 - 49. 4. de Guevara MS, Schauffele SI, Nicola-Siri LC, et al. Worsening of depressive symptoms 6 months after an acute coronary event in older adults is associated with impairment of cardiac autonomic function. J Affect Disord 2004;80:257 - 62. 5. Yeragani VK, Rao KA, Smitha MR, et al. Diminished chaos of heart rate time series in patients with major depression. Biol Psychiatry 2002;51:733 - 44. 6. Hall M, Vasko R, Buysse D, et al. Acute stress affects heart rate variability during sleep. Psychosom Med 2004;66:56 - 62. 7. Kop WJ, Verdino RJ, Gottdiener JS, et al. Changes in heart rate and heart rate variability before ambulatory ischemic events(1). J Am Coll Cardiol 2001;38:742 - 9. 8. Vrijkotte TG, van Doornen LJ, de Geus EJ. Effects of work stress on ambulatory blood pressure, heart rate, and heart rate variability. Hypertension 2000;35:880 - 6. 9. Hughes JW, Stoney CM. Depressed mood is related to highfrequency heart rate variability during stressors. Psychosom Med 2000;62:796 - 803. 10. Sheffield D, Krittayaphong R, Cascio WE, et al. Heart rate variability at rest and during mental stress in patients with coronary artery disease: differences in patients with high and low depression scores. Int J Behav Med 1998;5:31 - 47. 11. Lehofer M, Moser M, Hoehn-Saric R, et al. Major depression and cardiac autonomic control. Biol Psychiatry 1997;42:914 - 9. 12. Watkins LL, Blumenthal JA, Carney RM. Association of anxiety with reduced baroreflex cardiac control in patients after acute myocardial infarction. Am Heart J 2002;143:460 - 6. 13. La Rovere MT, Bigger Jr JT, Marcus FI, et al. Baroreflex sensitivity and heart-rate variability in prediction of total cardiac mortality after myocardial infarction. ATRAMI (Autonomic Tone and Reflexes After Myocardial Infarction) Investigators. Lancet 1998; 351:478 - 84. 14. Agelink MW, Majewski TB, Andrich J, et al. Short-term effects of intravenous benzodiazepines on autonomic neurocardiac regulation in humans: a comparison between midazolam, diazepam, and lorazepam. Crit Care Med 2002;30:997 - 1006. 15. Feighner JP. Overview of antidepressants currently used to treat anxiety disorders. J Clin Psychiatry 1999;60(Suppl 22):18 - 22. 16. Schroeder C, Tank J, Boschmann M, et al. Selective norepinephrine reuptake inhibition as a human model of orthostatic intolerance. Circulation 2002;105:347 - 53. 17. Tulen JH, Bruijn JA, de Man KJ, et al. Cardiovascular variability in major depressive disorder and effects of imipramine or mirtazapine [Org 3770]. J Clin Psychopharmacol 1996;16: 135 - 45. 18. Glassman AH, O’Connor CM, Califf RM, et al. Sertraline treatment of major depression in patients with acute MI or unstable angina. JAMA 2002;288:701 - 9. 19. McFarlane A, Kamath MV, Fallen EL, et al. Effect of sertraline on the recovery rate of cardiac autonomic function in depressed patients after acute myocardial infarction. Am Heart J 2001; 142:617 - 23. 20. Blanchard EB, Martin JE, Dubbert PM. Non-drug treatment for essential hypertension. Elmsford (NY): Pergamon Press; 1988.

American Heart Journal Volume 149, Number 6

21. Bernardi L, Porta C, Spicuzza L, et al. Slow breathing increases arterial baroreflex sensitivity in patients with chronic heart failure. Circulation 2002;105:143 - 5. 22. Bennaroch EE. Central autonomic network: functional organization and clinical correlations. Armonck (NY): Futura Publishing Company, Inc; 1997. 23. Lehrer P, Vaschillo E, Vaschillo B. Resonant breathing biofeedback training to increase cardiac variability: rationale and manual for training. Appl Psychophysiol Biofeedback 2000;25:177 - 91. 24. Wiens AN, Fuller KH, Crossen JR. Paced auditory serial addition test: adult norms and moderator variables. J Clin Exp Neuropsychol 1997;19:473 - 83. 25. Floras JS, Butler GC, Ando SI, et al. Differential sympathetic nerve and heart rate spectral effects of nonhypotensive lower body negative pressure. Am J Physiol Regul Integr Comp Physiol 2001;281:R468 - 75. 26. Kamath MV, Hollerbach S, Bajwa A, et al. Neurocardiac and cerebral responses evoked by esophageal vago-afferent stimulation in humans: effect of varying intensities. Cardiovasc Res 1998;40: 591 - 9. 27. Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas 1977; 1:385 - 401. 28. Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J Health Soc Behav 1983;24:385 - 96.

Nolan et al 1137.e7

29. Del Pozo JM, Gevirtz RN, Scher B, et al. Biofeedback treatment increases heart rate variability in patients with known coronary artery disease. Am Heart J 2004;147:E11. 30. Giardino ND, Chan L, Borson S. Combined heart rate variability and pulse oximetry biofeedback for chronic obstructive pulmonary disease: preliminary findings. Appl Psychophysiol Biofeedback 2004;29:121 - 33. 31. Lehrer PM, Vaschillo E, Vaschillo B, et al. Heart rate variability biofeedback increases baroreflex gain and peak expiratory flow. Psychosom Med 2003;65:796 - 805. 32. Cowan MJ, Kogan H, Burr R, et al. Power spectral analysis of heart rate variability after biofeedback training. J Electrocardiol 1991; 23(Suppl):85 - 94. 33. Lehrer PM, Vaschillo E, Vaschillo B, et al. Biofeedback treatment for asthma. Chest 2004;126:352 - 61. 34. Vaschillo E, Lehrer P, Rishe N, et al. Heart rate variability biofeedback as a method for assessing baroreflex function: a preliminary study of resonance in the cardiovascular system. Appl Psychophysiol Biofeedback 2002;27:1 - 27. 35. Thayer JF, Lane RD. A model of neurovisceral integration in emotion regulation and dysregulation. J Affect Disord 2000;61:201 - 16. 36. Kubota Y, Sato W, Toichi M, et al. Frontal midline theta rhythm is correlated with cardiac autonomic activities during the performance of an attention demanding meditation procedure. Cogn Brain Res 2001;11:281 - 7.