Last October, the 2009 Heart-Brain Summit—the fourth annual summit of this type presented by the Bakken Heart-Brain Institute—was held in Chicago and built on the the first three summits’ tradition of open-minded discussion, out-of-the-box thinking, scholarly activity, and engagement of attendees from varied backgrounds.
DEPRESSION AND HEART DISEASE: A WATERSHED YEAR, OR JUMPING THE GUN?
The year leading up to the 2009 summit may be remembered as a watershed period for the field of heart-brain medicine, in light of the American Heart Association’s (AHA’s) inclusion of the recommendation to screen patients with coronary artery disease (CAD) for depression in its science advisory on depression and CAD.1 As has been discussed at prior Heart-Brain Summits, there is incontrovertible evidence in the literature that CAD patients with depression have a worse prognosis than do their counterparts without depression.2–6 While the link is clear, the etiology or mechanism behind depression’s association with worse CAD outcomes is debated. Possible reasons for the association range from greater nonadherence with medical therapy7 to increased systemic inflammation related to the decreased vagal tone associated with depression.8 Furthermore, there is clear evidence that patients with depression and CAD can be treated for their depression safely with cognitive and pharmacologic therapy.5,9 What is lacking, however, is convincing data that the treatment of depression in patients with CAD leads to improved outcomes.10
The topic for the first half of the opening day of the 2009 summit was whether the AHA has gotten ahead of itself in its science advisory1 and whether we should require demonstrable benefits from the treatment of depression in CAD patients before screening for depression is recommended in all patients with CAD. This is a critically important question for the field as well as for the Bakken Heart-Brain Institute, which under our leadership has been advocating for a clinical trial to address this very issue. Cardiologists addressing this question were well reminded that logical therapeutic targets without proven end points have failed us in the past. For instance, it was a rational concept that the suppression of premature ventricular contractions in patients with a history of acute myocardial infarction would lead to decreased ventricular tachycardia and death. Unfortunately, when this concept was put to the test in a randomized clinical trial, increased death was observed in the treatment group.11 More recent examples—and perhaps more applicable to depression, given its chronic nature—come from recent clinical trials demonstrating that tight blood sugar control is associated with higher mortality than moderate blood sugar control in critically ill patients12 and that intensive blood pressure control does not yield greater reductions in cardiovascular events compared with moderate blood pressure control in patients with type 2 diabetes.13
So we are faced with a chronic disease state—depression—that is clearly linked to adverse outcomes and death in patients with CAD. In the context of this association, we also know the following:
- The AHA science advisory recommends that we screen all CAD patients for depression.
- Treating depression in heart disease patients is safe.
- There is no clear proof that treating depression will reverse the increased risk associated with depression in patients with CAD.
- There is a community of physicians who treat CAD patients who are skeptical about therapies that do not have outcomes data.
The summit’s first morning concluded with a debate on whether now is the time for a large-scale multicenter randomized trial, which raised several important issues:
- The limited effectiveness of treatment for depression (approximately 30% to 40%)
- The ethics of randomizing a patient with depression to placebo
- The required size of the trial, given the efficacy of antidepressant therapy
- Measures to define response to therapy
- The utility of surrogate markers for adverse events in CAD versus a mortality end point.
The discussion and presentations were excellent and animated. In the end, each attendee was left to reach his or her own conclusion. Personally, one of us (M.S.P.) was surprised to be left with the conclusion that we are not ready for a definitive clinical trial.
In the cardiovascular medicine literature we were faced with a similar situation regarding the management of patients with atrial fibrillation. In the AFFIRM trial, patients were randomized to conservative treatment (rate control and warfarin) or aggressive treatment (rate control, warfarin, and any and all therapies to convert to and maintain normal sinus rhythm).14 Ultimately there was no difference between the groups, with a trend toward improved outcomes in the conservatively treated patients. What we really learned was that our therapies to convert to and maintain normal sinus rhythm were inadequate, and that in the case of atrial fibrillation at least we could clearly identify which patients did not respond to therapy.14 These findings ultimately may have led the field astray, as we still do not know if we have efficacious therapies for the treatment of atrial fibrillation and whether patients would benefit.