Antidepressants might increase compliance with cardiovascular disease risk reduction interventions. However, antidepressants have been linked to deleterious metabolic effects. In the present multicenter study, we sought to determine whether patients who take antidepressants derive the expected benefits from cardiac rehabilitation in terms of improvements in multiple atherosclerotic risk factors. A cohort of 26,957 patients who had completed a baseline assessment before participating in an exercise-based cardiac rehabilitation program constituted the study population. The patients were stratified into 3 cohorts (i.e., nondepressed, depressed unmedicated, and depressed medicated) at baseline according to a self-reported history of depression and the current use of antidepressants. Risk factors were assessed at baseline and after ∼12 weeks of program participation. A self-reported history of depression was present at baseline in 5,172 patients (19.2%). Of these patients, 2,147 (41.5%) were taking antidepressants. Patients in the nondepressed cohort (49.4% completion) were more likely (p <0.001) to complete the exit assessment than patients in the depressed unmedicated (44.5% completion) or depressed medicated (43.5% completion) cohorts. Patients in all 3 cohorts who completed the exit assessment showed significant improvement in multiple risk factors. Moreover, the magnitude of improvement in blood pressure, serum lipids and lipoproteins, fasting glucose, weight, and body mass index was similar (p >0.05) in patients taking antidepressants and those who were not. In conclusion, our study is the first to show that antidepressants do not offset the average magnitude of improvement in multiple atherosclerotic risk factors that occurs with completion of a cardiac rehabilitation program.
Exercise-based cardiac rehabilitation (CR) is a well-accepted standard of care for patients with coronary heart disease (CHD), including those with depression. Antidepressant medications, especially selective serotonin reuptake inhibitors (SSRIs), are safe and moderately effective in helping to alleviate depression in patients with CHD and can also improve adherence to medical therapy. However, antidepressants can have deleterious effects on certain atherosclerotic risk factors and could potentially offset the magnitude of risk reduction that occurs with CR. Regarding the latter concern, although exercise training combined with antidepressants has a favorable effect on depressive symptoms, no data are currently available on the effect of CR on multiple atherosclerotic risk factors specifically in patients taking antidepressants. Therefore, in the present multicenter study, we sought to determine whether patients with a self-reported history of depression who were receiving treatment with antidepressants derived the expected benefits from CR in terms of improvements in multiple atherosclerotic risk factors.
Methods
A cohort of 26,957 consecutive patients who had completed a baseline assessment before participating in a 12-week, exercise-based, phase II, CR program conducted at 35 outpatient facilities in the United States from 1998 to 2009 constituted the study population. The outpatient facilities were distributed geographically throughout the United States (4 in the Northeast, 11 in the Southeast, 11 in the Midwest, 4 in the Southwest, and 5 in the Northwest); 10 were in urban, teaching hospitals and 25 were in community, nonteaching hospitals (11 of which were rural). All patients provided informed consent, and the Emory University (Atlanta, Georgia) institutional review board requirements were met.
In addition to supervised exercise training, the CR program included the use of standardized data collection procedures, computer-generated patient goals and action plan reports, written and audio patient education modules, and a computerized database, all made available as components of a commercial CHD risk reduction program, as previously described. During the baseline assessment, an extensive health history questionnaire, including information on medical history, medications, and atherosclerotic risk factors, was jointly completed by the program participants and staff. For those patients who completed the program, a modified version of the questionnaire was completed as a part of the exit assessment. The assessments also included measurement of height, weight, blood pressure at rest, and fasting total cholesterol, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein cholesterol, triglycerides, and glucose, although not all measurements were performed for all patients.
As a part of the baseline health history questionnaire, the patients were asked to review a list of medical conditions, including “depression,” and to identify from the list “all significant medical problems you have ever experienced at any time in your life, past or present.” For the purposes of the present study, the patients who identified depression as a medical problem were categorized as having a self-reported history of depression. Additionally, for each identified medical condition, the patients were asked to indicate whether it was “still currently a problem”; patients who indicated that depression was still currently a problem were categorized as having current self-reported depressive symptoms.
Using data from the baseline health history questionnaire, the patients were stratified into 1 of 3 study cohorts according to a self-reported history of depression and the current use of antidepressant medication, as follows: nondepressed cohort (no self-reported history of depression plus not currently taking antidepressant medication); depressed unmedicated cohort (self-reported history of depression but not currently taking antidepressant medication); and depressed medicated cohort (self-reported history of depression plus currently taking antidepressant medication).
For categorical variables, the statistical significance of differences between groups and changes from baseline was analyzed using chi-square tests. For continuous variables, the statistical significance of changes from baseline was analyzed using paired t tests, and the significance of differences between groups was analyzed using unpaired t tests. The tests were 2-sided, and statistical significance was established at p ≤0.05.
Results
Select baseline demographic and clinical characteristics of the study patients are summarized in Table 1 . A self-reported history of depression was identified at baseline in 5,172 patients. Of these patients, 2,147 (41.5%) were and 3,025 (58.5%) were not taking antidepressants. Of the 3,025 patients in the depressed unmedicated cohort, 1,649 (54.5%) had current self-reported depressive symptoms at baseline. Of the 2,147 in the depressed medicated cohort, 1,453 (67.7%) had current self-reported depressive symptoms at baseline (p <0.001 for depressed unmedicated vs depressed medicated cohort). The antidepressant medications taken by patients in the depressed medicated cohort included SSRIs (1,517 patients [70.7%]), serotonin-norepinephrine reuptake inhibitors (345 [16.1%]), norepinephrine-dopamine reuptake inhibitors (290 [13.5%]), tricyclic antidepressants (132 [6.1%]), serotonin antagonist and reuptake inhibitors (87 [4.1%]), noradrenergic and specific serotonergic antidepressants (49 [2.3%]), and monoamine oxidase inhibitors (1 patient). Of the patients taking antidepressants, 1,893 (88.2%) were receiving monotherapy.
| Variable | Value |
|---|---|
| Age (yrs) | 68.3 ± 11.6 |
| Gender | |
| Men | 69.8% |
| Women | 30.2% |
| Race | |
| Black | 5.6% |
| White | 90.2% |
| ≥1 Year of college education | 63.7% |
| Known atherosclerotic cardiovascular disease | 92.7% |
| Previous myocardial infarction | 40.3% |
| Previous coronary artery bypass grafting | 42.4% |
| Previous percutaneous coronary intervention | 44.0% |
| Previous stroke | 4.7% |
| Known diabetes mellitus | 24.0% |
| Self-reported history of depression | 19.2% |
| Smoke cigarettes | 6.7% |
| Systolic blood pressure (mm Hg) | 123 ± 18 |
| Diastolic blood pressure (mm Hg) | 71 ± 3 |
| Total cholesterol (mg/dl) | 168 ± 45 |
| Low-density lipoprotein cholesterol (mg/dl) | 96 ± 37 |
| High-density lipoprotein cholesterol (mg/dl) | 43 ± 14 |
| Triglycerides (mg/dl) | 155 ± 110 |
| Glucose (mg/dl) | 116 ± 40 |
| Weight (lb) | 193.0 ± 43.3 |
| Body mass index (kg/m 2 ) | 29.5 ± 5.9 |
Patients in the nondepressed cohort were more likely to complete the exit assessment than patients in either of the 2 depressed cohorts (exit assessment completion rate of 49.4% for the nondepressed cohort, 44.5% for the depressed unmedicated cohort, and 43.5% for the depressed medicated cohort; p <0.001 for nondepressed vs depressed unmedicated and depressed medicated cohorts; p = 0.514 for the depressed unmedicated vs depressed medicated cohorts). Patients in the nondepressed, depressed unmedicated, and depressed medicated cohorts who completed exit assessments did so after an average of 93, 107, and 112 days after the baseline assessment, respectively.
Select baseline demographic and clinical characteristics of the patients in the 3 study cohorts who completed both the baseline and exit assessment are summarized in Table 2 . The nondepressed cohort was older, included fewer women and cigarette smokers, and had lower total cholesterol, high-density lipoprotein cholesterol, triglycerides, weight, and body mass index than the depressed cohorts. With the exception of gender (greater percentage of women in the depressed medicated cohort) and diastolic blood pressure (slightly greater in the depressed unmedicated cohort), no statistically significant differences were present between the 2 depressed cohorts at baseline.
| Characteristic | Study Cohort | ||
|---|---|---|---|
| Nondepressed (n = 9,995) | Depressed Unmedicated (n = 1,345) | Depressed Medicated (n = 935) | |
| Age (yrs) | 69.9 ± 11.1 ∗ † | 67.9 ± 11.3 ∗ | 67.2 ± 10.5 † |
| Gender | |||
| Men | 74.8% ∗ † | 61.8% ∗ ‡ | 56.1% † ‡ |
| Women | 25.2% ∗ † | 38.2% ∗ ‡ | 43.9% † ‡ |
| Smoke cigarettes | 4.6% ∗ † | 7.2% ∗ | 8.0% † |
| Blood pressure (mm Hg) | |||
| Systolic | 123 ± 18 (n = 9,769) | 123 ± 19 (n = 1,319) | 122 ± 18 (n = 920) |
| Diastolic | 71 ± 3 ∗ (n = 9,763) | 72 ± 3 ∗ ‡ (n = 1,319) | 71 ± 3 ‡ (n = 920) |
| Cholesterol (mg/dl) | |||
| Total | 166 ± 45 ∗ † (n = 3,552) | 172 ± 42 ∗ (n = 476) | 172 ± 47 † (n = 329) |
| Low-density lipoprotein | 94 ± 36 (n = 3,455) | 98 ± 36 (n = 457) | 95 ± 38 (n = 310) |
| High-density lipoprotein | 42 ± 14 ∗ † (n = 3,528) | 44 ± 13 ∗ (n = 475) | 44 ± 15 † (n = 326) |
| Triglycerides (mg/dl) | 149 ± 101 ∗ † (n = 3,513) | 164 ± 135 ∗ (n = 474) | 170 ± 110 † (n = 323) |
| Glucose (mg/dl) | 115 ± 36 (n = 1,533) | 117 ± 42 (n = 224) | 117 ± 37 (n = 164) |
| Weight (lb) | 191.2 ± 41.1 ∗ † (n = 9,972) | 193.9 ± 44.5 ∗ (n = 1,345) | 195.4 ± 45.3 † (n = 935) |
| Body mass index (kg/m 2 ) | 29.1 ± 5.4 ∗ † (n = 9,972) | 30.1 ± 6.2 ∗ (n = 1,345) | 30.5 ± 6.2 † (n = 935) |
∗ Differences between nondepressed and depressed unmedicated cohorts were statistically significant (p ≤0.05) as indicated for the following: age, gender, smoke cigarettes, diastolic blood pressure, body mass index, p <0.001; triglycerides, p = 0.004; total cholesterol, p = 0.01; weight, p = 0.025; and high-density lipoprotein cholesterol, p = 0.033.
† Differences between nondepressed and depressed medicated cohorts were statistically significant (p ≤0.05) as indicated for the following: age, gender, smoke cigarettes, triglycerides, body mass index, p <0.001; weight, p = 0.003; total cholesterol, p = 0.022; and high-density lipoprotein cholesterol, p = 0.023.
‡ Differences between depressed unmedicated and depressed medicated cohorts were statistically significant (p ≤0.05) as indicated for the following: diastolic blood pressure, p <0.001; gender, p = 0.007.
The effect of the CR program on key clinical outcome measures in the patients from the 3 cohorts who completed the baseline and follow-up assessments is summarized in Table 3 . With the exception of fasting glucose levels in the patients in the depressed medicated cohort, statistically significant improvements were observed for all measures in the 3 cohorts. With the exception of a greater reduction in total and LDL cholesterol in the nondepressed versus the depressed unmedicated cohort, the magnitude of improvement was not statistically different among the 3 cohorts.
| Outcome Measure | Change from Baseline | ||
|---|---|---|---|
| Nondepressed | Depressed Unmedicated | Depressed Medicated | |
| Blood pressure (mm Hg) | |||
| Systolic | −3 ± 18 (n = 9,769; p <0.001) | −3 ± 19 (n = 1,319; p <0.001) | −3 ± 18 (n = 920; p <0.001) |
| Diastolic | −2 ± 3 (n = 9,763; p <0.001) | −2 ± 3 (n = 1,319; p <0.001) | −2 ± 3 (n = 920; p <0.001) |
| Cholesterol (mg/dl) | |||
| Total | −13 ± 39 ∗ (n = 3,552; p <0.001) | −8 ± 37 ∗ (n = 476; p <0.001) | −9 ± 40 (n = 329; p <0.001) |
| Low-density lipoprotein | −12 ± 33 ∗ (n = 3,455; p <0.001) | −8 ± 35 ∗ (n = 457; p <0.001) | −9 ± 36 (n = 310; p <0.001) |
| High-density lipoprotein | 2 ± 11 (n = 3,528; p <0.001) | 2 ± 11 (n = 475; p = 0.002) | 2 ± 11 (n = 326; p <0.001) |
| Triglycerides (mg/dl) | −15 ± 86 (n = 3,513; p <0.001) | −16 ± 108 (n = 474; p <0.001) | −14 ± 98 (n = 323; p = 0.012) |
| Glucose (mg/dl) | −6 ± 36 (n = 1,533; p <0.001) | −9 ± 39 (n = 224; p <0.001) | −2 ± 41 (n = 164; p = NS) |
| Weight (lb) | −1.7 ± 9.7 (n = 9,972; p <0.001) | −1.9 ± 10.5 (n = 1,345; p <0.001) | −1.6 ± 9.5 (n = 935; p <0.001) |
| Body mass index (kg/m 2 ) | −0.2 ± 1.5 (n = 9,972; p <0.001) | −0.3 ± 1.8 (n = 1,345; p <0.001) | −0.2 ± 1.7 (n = 935; p <0.001) |
∗ Differences between nondepressed and depressed unmedicated cohorts were statistically significant (p ≤0.05) as indicated for the following: total cholesterol, p = 0.005; LDL cholesterol, p = 0.015; no statistically significant differences were found between nondepressed and depressed medicated cohorts or depressed unmedicated and depressed medicated cohorts.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
