Transcript ABCT2009

Mediators of Combined Cardiorespiratory Biofeedback and Dialectical Behavioral Skills Therapy for Treating
Major Depressive Disorder and Low Heart Rate Variability Following Myocardial Infarction.
Milton Brown, Priya Chaudhri, and Richard Gevirtz
California School of Professional Psychology at Alliant International University, San Diego, CA
Cardiovascular disease (CVD) remains the leading cause of death and
morbidity for men and women in the United States. Recent studies
have shown that low heart rate variability (HRV) and depression
correlate with each other and predict death and adverse cardiac
events. 15-40% of CVD patients have depressive disorders. The
presence of major depressive disorder in patients with recent acute
myocardial infarction or unstable angina more than doubles the risk of
cardiac death in both men and women. Risk of death in CVD patients
with low HRV is four times higher than for CVD patients with high
HRV. In addition to low HRV, studies of depressed patients with CVD
have found evidence of elevated heart rate, exaggerated heart rate
responses to physical stressors, high variability in ventricular
repolarization, low baroreceptor sensitivity, and inflammation. Due to
the limited benefits from pharmacotherapy, it is critical that more
effective treatment options are found. There has been promising
preliminary evidence for cardiorespiratory HRV biofeedback in the
treatment of depression and low HRV, however it has yet to be
confirmed as an effective treatment option with post-myocardial
infarction patients. This biofeedback technique gives feedback about
fluctuations in HRV during diaphragmatic breathing to optimize the
rate and quality of breathing to maximize HRV and strengthen
homeostatic reflexes and vagal tone. Dialectical Behavior Therapy
(DBT; Linehan, 1993) is effective for emotion regulation, but only two
studies have verified that DBT improves depressive disorders, and
none have applied DBT to post-MI patients. This study is the first
randomized controlled trial to examine the effects of a brief DBTbiofeedback emotion regulation skills training intervention for post
myocardial infarction patients with major depressive disorder.
Mindfulness, emotion regulation, and heart rate variability were
examined as mediators of the treatment effect.
Hamilton Rating Scale for Depression
Mediation of Group Difference in HRSD Change
mediation
Covariates
Cohen’s d % reduction
p-value*
none
0.83
FFMQ
0.60
28%
.001
DERS
0.49
41%
.016
DERS+FFMQ 0.34
59%
35
30
25
20
15
10
DBT + biofeedback + sertraline
5
sertraline only
0
0
8
Pre-treatment
Post-treatment
12
Number of Weeks
Note: The thick lines indicate the HLM predicted scores (per group), and the thin lines indicate the means of the actual observed scores
Beck Depression Inventory
* Krull-MacKinnon mediation test
35
30
Results
25
● The DBT-biofeedback group had much larger decreases in
depression severity than the control group
• A higher proportion did not meet criteria for MDD at posttreatment (DISH, p<.001).
● Participants practiced biofeedback moderately and amount
of practice correlated with depression improvement.
● Contrary to our hypotheses:
• The DBT-biofeedback group did not consistently show
larger improvements in HRV
• HRV did not mediate the group difference in depression
improvement: SDNN-BDI (p = .614), SDNN-HRSD
(p = .613). VLF-BDI (p = .988), VLF-HRSD (p = .657).
● Measures of emotion regulation and mindfulness mediated
much of the group difference in depression improvement.
20
Method
15
Participants (N = 60):
■ age M = 56 years, 50% male, 50% Caucasian, 55% college grads
■ private practice cardiac rehabilitation clinic in California
■ current diagnoses: cardiovascular disease (CVD), prior myocardial
infarction (MI), Major Depressive Disorder (assessed via DISH)
10
Outcome Measures:
• Depression Interview and Structured Hamilton (DISH)
• Hamilton Rating Scale for Depression
• Beck Depression Inventory (BDI-II, 2nd Ed.)
Mediator Measures:
• Difficulties in Emotion Regulation Scale (DERS)
• Five Facet Mindfulness Questionnaire (FFMQ)
• Heart Rate Variability (HRV): Standard Deviation of Normal-Normal
Beats (SDNN) and Very Low Frequency (VLF)
Procedures:
Measures were given at pre-treatment, post-treatment (Week 8) and
follow-up (Week 12)
Treatment Group: Received 8 weeks of a once-per-week DBT skills
training group (90 minute sessions) focused on behavioral activation,
mindfulness, and cardiorespiratory biofeedback emotion regulation
training, combined with sertraline antidepressant treatment . The
biofeedback training involved practice of slow diaphragmatic
breathing (about 5-6 breaths per minute) with the aide of a
StressEraser (Helicor Inc.), a handheld HRV biofeedback device.
Goal was 20 minutes and 50 points per day.
Control Group: Sertraline antidepressant treatment
Data Analyses: Hierarchical linear modeling (HLM). The Krull &
MacKinnon (2001) method (non-lagged) was used to test mediation.
Mediation of Group Difference in BDI Change
mediation
Covariates
Cohen’s d % reduction
p-value*
none
1.12
FFMQ
0.91
19%
.002
DERS
0.68
39%
.0004
DERS+FFMQ 0.59
47%
5
0
0
8
Pre-treatment
Post-treatment
12
Number of Weeks
Primary Outcomes
Group Difference in Depression Improvement*
Outcome
HRSD
BDI
HRV-SDNN
HRV-VLF
Cohen’s d
0.83
1.12
0.40
0.67
* Treatment X time interaction effect
p-value
.001
.0001
.409
.006
Conclusions
● DBT emotion regulation skills training combined with
cardiorespiratory biofeedback and sertraline medication
may be significantly more effective for cardiac rehabilitation
patients with depressive disorders than sertraline
medication alone.
● The emotion regulation and HRV biofeedback skills training
may be effective because it effectively improves emotion
regulation and mindfulness.
● Although we found no evidence of the intervention working
because of enhanced HRV, it is possible that biofeedback
improves functioning of baroreceptors, vagal afferent
pathways, and ultimately serotonin levels.
● More rigorous biofeedback training may be necessary.