Observational study suggests SSRI antidepressants may increase mortality in CAD patients

Shelley Wood

March 04, 2006

Denver, CO - Controlling depression with drugs, and especially selective serotonin-reuptake inhibitors (SSRIs), may increase the risk of dying from cardiovascular disease, a new observational study suggests[1]. Duke researchers, led by Dr Lana Watkins, called the findings "unexpected," particularly since depression is an established risk factor for worse outcomes in patients with cardiovascular disease and prescription of antidepressants has increased partly out of the belief that improving depressive symptoms could also improve survival.

Dr Lana Watkins
(Source: Duke University Medical Center)

As well, results from the SADHART study, reported by heart wire , indicated that the SSRI sertraline was not associated with an increase in surrogate measures of adverse cardiovascular effects but was helpful in treating depression.

Watkins presented the results from the observational study at the American Psychosomatic Society 64th Annual Meeting in Denver, CO.

"This finding that antidepressant use was an independent risk factor for mortality in patients with coronary artery disease was quite unexpected," Watkins commented in a press statement. "We were surprised, since antidepressants, particularly the newer class of antidepressants known as SSRIs, have been generally considered safe; however, even after taking into account many patient variables, as well as the type of antidepressant, the risk still remained. So there is something important going on here that we don't fully understand."

Data on SSRI use and mortality was collected as part of a prospective study examining the effects of anxiety and depression on risk of events in patients hospitalized to undergo coronary angiography. Out of 922 patients, 179 were taking some type of antidepressant medication, compared with 742 not on antidepressants. Compared with patients not on medication, those taking antidepressant drugs had, on average, significantly fewer diseased vessels and significantly higher ejection fractions but were significantly more likely to have other medical comorbidities. Over follow-up, antidepressant users were twice as likely to die as people not on antidepressant medication over a median of three years, even after age, gender, ejection fraction, smoking history, comorbidities, education, and depression score were controlled for. SSRIs were the most common type of antidepressant medication being taken, with more than two thirds of patients taking them. Mortality was higher in patients taking SSRIs compared with non-SSRIs, but this difference was not statistically significant.

Antidepressant use

Factor No antidepressant use Antidepressant use p
BDI score* 7 11 <0.001
Vessels with significant CAD (n) 2.1 1.8 <0.05
Ejection fraction (%) 54 56 <0.05
Advanced or multiple medical comorbidity (%) 23 32 <0.01
Death during follow-up (%) 12.5 21.4 <0.01
*Beck Depression Inventory score; increasing score=worse depression

"The present findings suggest that antidepressant use during hospitalization for coronary angiography may be predictive of increased risk of mortality," the authors conclude.

"Future studies are needed to identify whether the increased risk attributable to antidepressant use is secondary to other variables such as worsening depression or worsening medical disease during follow-up."

While the further research is required, a number of hypotheses could explain the link between antidepressants and increased mortality disease, Watkins notes, including effects on the immune system, blood platelets, body weight, insulin resistance, blood pressure, or lifestyle factors such as smoking, alcohol use, and physical-activity level.

Commenting on the study for heartwire , Dr Robert Carney (Washington University School of Medicine, St Louis, MO) called the findings "unexpected" and emphasized that further studies are needed.

"Some studies, although only one randomized clinical trial, have suggested that not only are SSRIs safe, but they may have cardiac benefits," he explained. "The only randomized clinical trial of SSRIs in post-MI patients is SADHART, published in JAMA a few years ago. [The SADHART investigators] found the drug to be safe, with a suggestion of a beneficial effect on cardiac outcomes," although the trial was not adequately powered to detect such a difference. 

"It also occurs to me that perhaps being on an antidepressant in this study was a marker for having had more severe depression, over a longer period," Carney added. "Therefore, the effect of depression was not so much a reflection of the current level, as measured by the BDI score, but of their overall exposure, perhaps over many years."

Also commenting on the study to heart wire , Dr Alexander Glassman (Columbia University, New York, NY) pointed out that two other "more or less randomized" studies in addition to SADHART also examined the relationship between treatment for depression and the heart, but one study looked at psychotherapy, not drug use, while the other was conducted in ischemic-stroke patients, not CAD patients. All three studies, however, suggested a benefit of depression treatment.

"To be honest, given the available data I doubt the Duke study is correct," Glassman commented. "However, it will raise questions, and the real answer is that we need a large well-controlled, randomized trial examining whether SSRI treatment reduces the well-documented increase in cardiovascular risk seen with depression following any ACS event."

Watkins agrees. "The bottom-line is that more research is necessary," she told heartwire . "At this point, there is insufficient evidence to result in a shift of clinical practice."  

Duke investigators will soon be launching a clinical trial comparing the cardiovascular effects of exercise and SSRIs in the treatment of depression, which may help clarify the findings from this observational study, the press release notes.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.