Late life depression and cardiovascular autonomic function:
Major depressive disorder in the elderly (in those \>60 years of age), also known as late life depression (LLD), is common, disabling and associated with a high mortality rate caused by suicide as well as cardiovascular events compelling appropriate treatment. Research shows that in a naturalistic setting response rate to at least one antidepressant trial of adequate dose and duration alone is around 30-40% thereby necessitating usage of additional interventions. Such therapies include psychological therapies, and, recently, various forms of treatments loosely defined as mind-body therapies such as biofeedback, energy healing, meditation, guided imagery, and yoga. Mind-body therapies are being increasingly embraced by patients as they have negligible side effects, are easy to administer and display beneficial effects on the quality of life as well as comorbid anxiety. There is increasing research on the mechanisms and benefits of such therapies, however, good quality trial data is scant. It is well established that antidepressants work by mostly modifying neurotransmitter levels in the brain. On the other hand, some mind-body therapies target multiple organ systems and hence could offer neurobiological advantages as depression is now recognised as a multi-system disorder. This is particularly relevant to the LLD population where there is increased prevalence of comorbid cardiovascular disorders. Hence, some mind body therapies when offered in combination with antidepressants might have a beneficial effect on both depression and the cardiovascular system.
One of the ways of assessing the cardiovascular system is through measurement of various autonomic parameters i.e. heart rate, blood pressure and heart rate variability. The most commonly reported is heart rate variability (HRV) which is a manifestation of the interplay of the central nervous system and the autonomic nervous system on a beat-by-beat basis. One of the ways of its estimation is by calculating the elapsed time between two consecutive waves, called R waves, on a person's electrocardiogram (ECG). A consistent finding has been that there is reduced HRV in people who have suffered a myocardial infarction (MI), and this phenomena is a predictor of subsequent cardiac arrhythmia and even death. In addition, through previous work the PI has found that in LLD there are significant cardiovascular autonomic disturbances compared to age matched controls after controlling for various risk factors. Other studies conducted in depression across the human life span have found similar results. Hence, if there is a mind-body treatment that specifically targets autonomic dysfunction and has a positive benefit on depressive symptoms, it would be valuable to asses. The investigators would like to investigate such a therapy which has been selected based on previous reports of beneficial effects but has not been evaluated in a randomised controlled trial of LLD patients.
Automatic self-transcending meditation (ASTM):
ASTM is a class of meditation that helps quiet the mind and induces physiological and mental relaxation whilst the eyes are shut. It utilizes a specific sound value (mantra) to draw attention inward and permit the mind to experience a restful but alert state of consciousness. Research suggests that ASTM is easier to learn and to teach in comparison to other meditation techniques including mindfulness. Studies of adult and elderly ASTM practitioners have documented reductions in depressive symptoms, as well as improvements in cardiovascular function among elderly with and without cardiovascular disease. A study of adults with CVD further demonstrated improvements in HRV. Research further suggests ASTM may be particularly well suited to elderly populations. In a randomized controlled trial of elderly retirement home residents which evaluated ASTM with two other meditative techniques and treatment as usual, ASTM produced significantly greater improvements in cognitive function, cardiovascular function and quality of life than all other treatment conditions. A subsequent meta-analysis of all-cause mortality rates among hypertensive elderly who had participated in stress reduction interventions found that ASTM practitioners had a 30% lower cardiovascular mortality rate than four other meditative or relaxation interventions.
The investigators predict that ASTM augmentation is an effective intervention that ameliorates the autonomic disturbance associated with LLD, and possibly has beneficial effects on depressive symptoms as compared to a control treatment as usual (TAU) group.
Primary hypothesis: The investigators hypothesise that in patients with LLD, ASTM+TAU will lead to a significant increase in HRV from baseline to end of study period as compared to TAU.
Secondary hypotheses: The investigators hypothesise that ASTM+TAU will cause a) significant fall in depression scores b) significant improvement in other autonomic parameters including heart rate and blood pressure changes in presence of a physiological stress test (hand grip) c) significant improvement in depression related symptom pathologies including impaired quality of life and anxiety; compared to TAU.
Study Design:
This study is a single-centre, single blind longitudinal randomized controlled naturalistic trial. Research participants will be 96 men and women (48 in each group), 60-85 years of age, who have mild to moderate major depressive disorder (MDD). Participants will be recruited from primary, secondary and tertiary care centres in London, Ontario.
Study recruitment:
It is expected that the study will recruit from primary secondary and tertiary care practices in London identified by the investigators at a rate of at least 2 participants per week over a period of 76 weeks allowing attainment of a sample size of n=96. Advertisements about the study will be placed at key areas around the city including various community centres and libraries.
Usual standard of care at the trial site:
The standard of care for patients with late life depression involves starting them on an appropriate antidepressant from various classes such as SSRIs, (fluoxetine, fluoxamine, paroxetine, citalopram, escitalopram, sertraline, paroxetine), or NaSSA (mirtazapine) or Bupropion. Patients are also prescribed additional anxiolytic medications like benzodiazepines and trazodone if deemed necessary. Participants will be initiated and maintained on appropriate dosages of such medications as part of standard of care. The psychiatrist and/or his multidisciplinary team members might offer supportive therapy as part of standard of care.
Screening and initial assessments:
Potential participants will be screened as per inclusion and exclusion criteria. Upon selection, the following scales will be administered: interview for comorbid medical conditions (Cumulative illness rating scale- Geriatrics, CIRS-G), a screening cognitive examination (Mini Mental State Examination (MMSE)), depression severity assessment (Hamilton Rating Scale for Depression score (HRSD 17 items), Clinical Global Impression (CGI), self rated Geriatric Depression Scale (GDS)), anxiety (Geriatric Anxiety Inventory (GAI)), side effects (Toronto Side Effects Scale, (TSES)), Quality of Life (QOL profile seniors version (QOLPS), physical activity assessment (Physical activity scale of the elderly, PASE questionnaire).
Randomization:
Participants will be randomized to either ASTM+TAU or TAU equally (1:1) using computer generated randomisation numbers available at random.org. Concealment of randomisation will be ensured by independent staff performing randomisation.
Medication adherence follow-up:
All medication types will be permitted in this study. Any dosage modifications will be recorded.
Measurement of outcomes:
PRIMARY (HEART RATE VARIABILITY): Participants from both arms will be asked to report to the Laboratory for Brain and Heart Health at Western University Campus. Participants in both study arms will be requested to attend appointments at weeks 0 and 12 while participants in the ASTM arm will be requested to attend an additional assessment at 24 weeks.
1. Heart rate will be monitored using an electrocardiogram (ECG) with three adhesive leads.
2. Blood pressure will be measured using a small cuff placed on a finger (Finometer) or using a wrist cuff device (Colin Pilot). These continuous measures of blood pressure will be confirmed against values obtained periodically by an automated sphygmomanometer (Dinamap).
3. Electrocardiogram: A standard electrocardiogram will be collected via small surface electrodes on the chest to determine heart rate.
4. A bellows placed around the chest will provide information on respiratory excursions.
SECONDARY OUTCOME MEASURES:
Depression and comorbid symptoms will be assessed by a blinded rater on various scales at study visit days (week 0, 4, 8, and 12 for all participants and ASTM participants will attend an additional assessment at week 24). These scales are the rater assessed (Hamilton Rating Scale for Depression score (HRSD 17 items)) and Clinical Global Impression (CGI). Self rated scales include the Geriatric Depression Scale (GDS)), anxiety (Geriatric Anxiety Inventory (GAI)), Physical activity (PASE), adverse events (TSES) and quality of life (QOLPS).
Statistical analysis plan:
The main outcome of interest, HRV, will be calculated by standard deviation of all R-R intervals (SDNN) on ECG, root-mean square of successive differences (RMSSD), and number of R-R intervals differing by \>50 m sec from adjacent intervals (NN50) in time domain analysis.
Analyses will be conducted on a PC running SAS v9.4 (SAS Institute Inc. Cary, USA) on a Windows 7 Platform. Demographic and clinical characteristics of the intervention (SSM) and control (TAU) condition will be compared by independent samples t-tests and χ2 or Fisher's exact test for continuous and categorical data, respectively. Linear mixed models (described below), controlling for baseline score, will be used to compare the SSM and TAU groups' change score from baseline to 12-week follow-up on SDNN and LF HRV (primary outcome) and depression severity (HRSD-17) (secondary outcome). Other exploratory outcomes will be evaluated in a similar way. Linear mixed models will also be used to compare the percent change in HRSD-17 scores, whereas generalised linear models will be used to compare the proportion of patients who responded to the intervention (≥50% decrease from baseline on the HRSD, defined a priori) and the proportion of patients who achieved remission (scores ≤7 on the HRSD-17, defined a priori) at the end of intervention (week 12). Linear mixed models with variance components covariance structure will be used in comparing HRSD-17 scores between groups (SSM and TAU), time (week 0, 4, 8 and 12) and group × time interaction. All HRV-related measures will be natural log-transformed. Cohen's d will be calculated by dividing the mean difference in change scores between the two groups by the s.d. of the control groups' change score. Following established guidelines, the linear mixed models will account for the partially nested design, which allows clustering in the intervention group (because of participants experiencing a group-based intervention, added as a random effect) and no clustering in the control group. Using this method, different hierarchical structures can be fit for the two groups. PROC MIXED will be used for continuous outcome variables, and PROC GLIMMIX will be used for binary outcome variables. Restricted maximum likelihood estimates will be used and the d.f. adjusted with the Satterthwaite method. Residual and influence diagnostics will be checked to confirm if the models fit the data well. The level of statistical significance will be set at α = 0.05. Complete-case analysis will be used.
Interim analysis may be completed in order to meet one or more of these conditions a) funder reporting requirements and/or b) research training requirements of students and/or c) data safety monitoring for stop or continue decisions.
