More than metaphor links hearts and minds, but is the relationship between
depression and heart disease one of cause or coincidence?
Epidemiology shows a clear association. Between one-fifth and one-third of
people with heart disease also have depression, and depression predicts
development of cardiovascular disease in healthy people and increased
mortality in patients who have had heart attacks. Even those with depressive
symptoms but not clinically diagnosed major depression did worse than those
without such symptoms.
However, behavioral scientists and cardiologists are still searching for
definitive studies to prove causation and show that treating depression can
reduce cardiovascular mortality.
In short, even after years of research, the connection remains suggestive
but shadowy, said Nancy Frasure-Smith, Ph.D., of McGill University and the
University of Montreal, in a lecture at APA's 2005 annual meeting in Atlanta.
That ambiguity suggests that concern for depression rates low among
"Depression is off the radar of cardiac care," said
Cardiology has indeed been slow to acknowledge depression's role in
cardiovascular disease, agreed Christopher O'Connor, M.D., a professor of
medicine and director of the heart-failure program at Duke University Medical
Center, in an interview. Depression shows up rarely in the cardiological
literature, and cardiologists rarely screen for it, he said.
"Cardiologists think that when patients come in with cardiac events,
some depression is normal and that the patient will eventually get
better," O'Connor told Psychiatric News. "This is partly
true, but depressive symptoms in the hospital are a risk factor even if the
patient returns to baseline two weeks later."
Connecting depression and heart disease evolved in stages. Work by several
researchers in the 1970s showed a link between life stress and sudden death.
Frasure-Smith later analyzed results of an intervention trial testing a
distress-reducing intervention. There was no difference in outcomes between
the two arms, but she found that distress levels in the usual-care group
predicted cardiac mortality.
"The symptoms of distress that were most linked to prognosis were
complaints of poor concentration, feelings of unhappiness, not enjoying
activities, and other symptoms that seemed to be related to depression,"
she said. "This led my colleague François Lespérance and
me to look specifically at the importance of depression, but we were not the
first to consider the importance of depression."
She credits others, including Robert Carney, Ph.D., and Kenneth Freedland,
Ph.D., of the Department of Psychiatry at Washington University in St. Louis
with first concluding in the late 1980s that "major depressive disorder
is an important independent risk factor for the occurrence of major cardiac
events in patients with coronary artery disease."
Depression's relationship to heart disease meets many of the criteria for
causality laid out by Austin Bradford Hill, Ph.D., D.Sc., said Frasure-Smith,
but it has been hard to discern cause and effect in either direction. A
systematic review she conducted with Lespérance found 64 prospective
studies that used recognized measures of depression and included objective
recorded outcomes. They noted problems with sample sizes and characteristics
and with definitions of depression and outcome measures.
"Despite these issues," they wrote in the May-June supplement
to the journal Psychosomatic Medicine, "the bulk of the data
from protective studies with recognized indices of depression and objective
outcome measures is supportive of depression as a cardiac risk
In their own studies of heart attack and stable angina patients, she and
Lespérance found that depressed patients are more seriously ill than
other heart patients and that depression predicts mortality independently of
ejection fraction, hypertension, and heart failure severity.
Large, long-term studies are needed to determine whether depression causes
or worsens heart disease, said Frasure-Smith at APA's annual meeting. Any
valid explanation would also have to demonstrate biologically plausible
mechanisms by which depression caused heart disease and demonstrate that
treating depression reduced morbidity and mortality.
Many hypotheses have been put forth for how depression might harm the
cardiovascular system, although none is proven, said Frasure-Smith. Some have
suggested behavioral influences: depressed people are less likely to get
treatment or take their medications, for example.
Others have hypothesized a physiological connection by way of the autonomic
nervous system, which controls heart rate. Depressed people often have higher
heart rates, and a higher resting heart rate increases risk of sudden cardiac
death. Also, depressed patients have lower heart-rate variability, which also
carries an increased risk of death. Indicators of platelet activity and
inflammation are higher among depressed individuals. Researchers have noted
improved mood in patients taking lipophilic statin drugs, which reduce some
inflammatory marker levels.
Studies have found that people who eat more fish containing omega-3 fatty
acids have lower incidence of major depression. Omega-3 fatty acids are also
antiarrhythmic and antithrombogenic and increase heart-rate variability.
Cardiologists are more attuned to physiological than behavioral
explanations, said John Rumsfeld, M.D., Ph.D., an associate professor of
cardiology at the University of Colorado Health Sciences Center and the Denver
Veterans Affairs Medical Center.
"To date, cardiologists have embraced hypertension, diabetes, and
cholesterol levels as established risk factors and now treat them as part of
the standard of care," said Rumsfeld in an interview. "But
depression has remained wholly separate from the cardiovascular world. There's
no formal recommendation for when and how to screen for depression."
Rumsfeld said he is starting to screen patients using the Patient Health
Questionnaire-9 (PHQ-9) in his office waiting room. He argued that treating
depression can be justified to improve the patient's quality of life and to
improve adherence to cardiac-care recommendations, even in the absence of
evidence of benefits to cardiovascular physiology.
Only two treatment trials have attempted to solve the final part of the
puzzle— whether treating depressed heart patients for depression affects
"We don't know yet whether treating depression will reduce the
cardiac-event rate, which is why cardiologists haven't jumped on board,"
said O'Connor. They may also have had bad experiences in the 1980s when they
prescribed tricyclic antidepressants, which are now known to be
SADHART (Sertraline AntiDepressant Heart Attack Trial) tested the SSRI
sertraline against placebo but found no significant differences in outcome
over the 24 weeks of the trial. The study was designed as a safety test and
was too small and too short to draw useful conclusions, said
However, every site in SADHART had both a cardiologist and a psychiatrist
on the study team—a first, said O'Connor, who worked on the trial.
The ENRICHD (Enhancing Recovery in Coronary Heart Disease) study found that
patients who got cognitive-behavioral therapy and were followed for a mean of
29 months recorded a two-point difference on the Hamilton Depression Rating
Scale. That difference was statistically significant but there was no impact
on survival, said Frasure-Smith.
A recent secondary analysis of the ENRICHD data concluded that treating
depression reduced risk of death or recurrent heart attack in patients who had
already had a heart attack. After adjustment for baseline depression and
cardiac risk, use of SSRI antidepressants was associated with a 43 percent
reduced risk of all-cause mortality and death or recurrent heart attack,
according to a report by C. Barr Taylor, M.D., of Stanford Medical Center and
colleagues in the July Archives of General Psychiatry. They
cautioned, however, that their analysis was post hoc and observational, based
on patient self-report of SSRI dosage and use, and may have been confounded by
use of cognitive-behavioral therapy in half the patients.
"The results of this substudy are suggestive, but far from being
confirmatory," said Frasure-Smith in an interview. "Without
randomization, there is no way of assuring that the people with and without
SSRIs were identical in important background factors that may have influenced
outcome. Another randomized study would be needed to assess whether SSRIs
really causally influence cardiac events."
More than one large, well-designed trial will be needed to prove to both
psychiatrists and cardiologists that treating depression can save heart
patients' lives, she said.
"If any one treatment improved prognosis in coronary heart disease,
we would know it was due to the therapy but not necessarily that was due to
reducing depression," she said. "But if trials of several
different therapies improved outcomes then we'd be closer to knowing that
changing depression changes risk."
Planning is under way for a new version of SADHART that would recruit 3,000
to 4,000 patients with major and minor depression and test the effects of
cognitive-behavioral therapy and sertraline on cardiovascular outcomes, said
O'Connor. The Veterans Affairs medical system will soon begin a push to
recognize and treat depression in its disease-management program for heart
failure, said Rumsfeld.
All of that is good news to Frasure-Smith, who would like to see better
coordination between psychiatrists and cardiologists as they plan and run
"There is little evidence from clinical trials that we can change
depression in coronary artery disease patients, but there is enough evidence
to support additional trials," she said. "Trials to assess mood
should judge other cardiovascular factors, and trials of cardiovascular
treatments should assess mood."
More information is available online in "From Feeling Blue to
Clinical Depression: Exploring the Pathogenicity of Depressive Symptoms and
Their Management in Cardiac Practice" at<www.psychosomaticmedicine.org/cgi/content/full/67/Supplement_1/S2>
and "Reflections on Depression as a Cardiac Risk Factor" at<www.psychosomaticmedicine.org/cgi/content/full/67/Supplement_1/S19>.▪