Can Antidepressants Lower Coronary Disease Mortality?

Numerous studies have demonstrated that adding depression to existing cardiovascular disease significantly increases patients’ chances of dying within five years of diagnosis. However, new research presented in May at APA’s 2002 annual meeting indicates that adequate treatment of the patients’ depression may significantly improve their prognosis.

The apparent link between heart disease and depression appears to be a two-way street. Recent studies have provided compelling evidence that otherwise healthy, middle-aged individuals with depression have an increased risk of developing ischemic heart disease compared with individuals without depressive symptoms. Indeed, one recent study documented that in some individuals even “mental stress”—not necessarily anxiety or depression—can trigger decreased blood flow to the heart, leading to ischemia. Individuals with stress-induced ischemia were three times more likely to die within five years compared with people without ischemia brought on by mental stress, the study concluded.

Similarly, those individuals who have known coronary artery disease (CAD) and suffer a heart attack or undergo coronary artery bypass surgery and then subsequently develop depression have been shown to have a worsened prognosis compared with those who do not become depressed, according to a review of related research. Studies have reported that around 20 percent of persons with coronary artery disease have depression that meets DSM criteria for major depressive disorder, and as many as 30 percent to 40 percent of patients with CAD have less-severe symptoms. Recently, researchers reported that in a study of more than 900 people with CAD, followed for a minimum of nine years, patient scores on the Beck Depression Inventory (BDI) were predictive of mortality over a five-year period. Anxiety, anger, and social support were found not to be predictive of mortality, but subjects with higher BDI scores had an increased risk of death that was 3.5 times greater than that of the subjects with low BDI scores.

“The crucial remaining question,” said Alexander Glassman, M.D., a clinical professor of psychiatry at Columbia University, “is whether treating depression after a heart attack or stroke can decrease the risk of dying” from that event.

SADHART Established

Several years ago, Glassman and his colleagues set out to attempt to answer that question with SADHART—the Sertraline Antidepressant Heart Attack Randomized Trial.

“Our biggest initial problem was figuring out how large a sample size we’d need in order to show any effect,” Glassman said at the annual meeting session. They wanted to establish the safety and efficacy of using an antidepressant in cardiovascular disease. To do so with enough power to detect any statistically significant differences, “we were looking at enrolling an absolutely huge number of subjects.” The group settled on enrolling 200 patients in each arm of the prospective study, one group treated with sertraline (Zoloft) and the other treated with placebo.

“Previous work had been done with stable disease and things like Paxil and Prozac,” Glassman noted, “but nobody had done anything with unstable cardiovascular disease.”

Glassman coordinated a multicenter trial that eventually screened 11,500 patients who had suffered an acute myocardial infarction (MI). Slightly more than 3,300 met eligibility requirements with some symptoms of depression. The subjects had not sought treatment for depression and had generally mild symptoms. However, around 1,800 of those met criteria for major depression, and slightly fewer than 900 met the full inclusion/exclusion criteria. In the end, 186 were randomized to sertraline, and 183 were randomized to placebo for 26 weeks.

“There you see the relevance of statistical empowerment,” Glassman said. “In order to end up with 200 patients in each arm of the study, we had to screen nearly 12,000.”

The study used a measure of cardiac-pumping power to follow cardiac function following the MI and DSM criteria for depression to follow the patients’ major depression.

The results, now available in preliminary form, Glassman said, were interesting—not quite what he expected, but very significant. Publication of the complete data from the trial is expected around the end of the year.

Three Groups Assessed

The researchers broke the subjects into three groups for analysis in each arm of the trial: those patients who experienced their first episode of depression following their MI, those patients who had one prior episode of depression, and those patients who had two or more episodes of depression that the researchers categorized as “severe” depression.

“Overall, response rates [for depression] were 66 percent for those taking sertraline and 53 percent for those on placebo,” Glassman said. “But what is very interesting is when you look at response rates within the three different groups.” For those patients with their first episode of depression, there was no statistically significant difference in the response rates between sertraline and placebo (59 percent versus 55 percent, respectively). With one prior episode of depression, the difference was very significant, 72 percent for those on sertraline versus 51 percent for placebo. And for the “severe” category, 78 percent of those taking sertraline were responders compared with 45 percent of those taking placebo.

“These results, though, were really tied to the recurrence of the depression, not necessarily to the severity of depression,” Glassman noted. There were not statistically significant differences between higher versus moderate scores on the Hamilton Rating Scale for Depression.

The adverse-event profile also varied between the placebo and sertraline group, consistent with known side effects of the drug. The most common side effects were nausea and diarrhea.

Interestingly, the cardiac-complications profile also differed between the groups. “It looked like those taking sertraline had decreased risks for most major cardiac events,” Glassman said, including risk of having another MI, arrhythmias, or sudden cardiac death. However, while these differences did not reach statistical significance, there was a consistent trend for those taking sertraline to have lower rates of death within the study period, as well as lower rates of severe chest pain, arrhythmias, and stroke.

“I suspect that if we had a larger patient base, we might have seen these reach statistical significance,” Glassman noted, “but then we would have to have screened maybe 50,000 or 100,000 patients.

“In the end, what we saw was that sertraline was indeed safe and effective at treating recurrent depression following an MI, but not in first-episode depression.” ▪