NEW ORLEANS – The addition of coronary bypass surgery to aggressive medical care failed to reduce the primary end point of all-cause death in patients with coronary artery disease and heart failure in the Surgical Treatment for Ischemic Heart Failure trial.
Still, several experts characterized the trial as a success and its lead author, Dr. Eric Velazquez, said the data support coronary artery disease (CAD) assessment in all patients presenting with heart failure.
After 6 years of follow-up among 1,202 randomized patients, there was a nonsignificant reduction in Kaplan Meier all-cause mortality rates with coronary artery bypass grafting (CABG) plus medical therapy, from 46% to 41%, Dr. Velazquez reported on April 4 at the annual meeting of the American College of Cardiology.
After adjusting this outcome for prespecified baseline variables, the hazard ratio was 0.82 and P value .039.
Crude all-cause mortality rates fell, albeit not significantly, from 41% to 36%, as simultaneously reported online (N. Engl. J. Med. 2011 [10.1056/NEJMa1100356]).
Among key secondary outcomes, however, bypass surgery significantly reduced Kaplan-Meier cardiovascular mortality event rates, from 39% to 32%, and crude cardiovascular mortality rates, from 33% to 28%.
Both Kaplan-Meier event rates and crude rates of death or cardiovascular hospitalization were significantly reduced with bypass, from 68% to 58%. The secondary outcomes remained significant after adjustment.
As anticipated, CABG was associated with an early risk of death that took 2 years to abate, observed Dr. Velazquez, director of the cardiac diagnostic unit and echocardiography laboratories at Duke University Medical Center in Durham, N.C.
"Decision making for CABG is complex," he said. "It should be individualized and now with the results of the STICH trial, patients should be informed of the short-term risk for a potential long-term benefit."
In all, 17% of the 602 patients randomly assigned to medical therapy alone crossed over to receive bypass surgery before the end of follow-up, and 9% of the 610 patients assigned to CABG received medical treatment only.
When the data were analyzed on an as-treated basis from the resulting 592 medical therapy and 620 CABG patients, the addition of bypass surgery significantly reduced deaths from any cause by 11%, from 49% to 38% (P value less than .001; HR, 0.70), Dr. Velazquez said.
The researchers then performed a per-protocol analysis of the 537 medical therapy patients who did not cross over to CABG during the first year of follow-up and the 555 CABG patients who actually underwent the procedure. Once again, the primary outcome of all-cause mortality was significantly reduced 11%, this time from 48% to 37% (P = .005, HR 0.76).
Current guidelines don’t support evaluation of coronary artery disease in patients with heart failure who present without chest pain, resulting in a lost opportunity for clinicians and leaving the exact number of patients for whom the results of STICH would apply unclear, Dr. Velazquez said in an interview.
"I think the guidelines need to recognize that coronary artery disease presents in many ways in our patients and that evaluation of coronary artery disease is important not only for consideration of bypass surgery, but also to optimize medical therapy and CAD medication," he said.
Despite the medical adherence and operative results achieved in the trial, STICH-like patients remain at substantial risk with a 5-year mortality rate of 40% with medication only.
Invited discussant Dr. Bernard Gersch said "We have known for decades that, in patients with left-ventricular dysfunction and ischemia, left-ventricular dysfunction is the major cause of mortality. And you have now proven the concept. This is an incredible trial. It’s a stunning achievement and very difficult to do."
Future analyses of the mechanisms of benefit associated with bypass surgery will prove important in determining whether the benefit is from an improvement in diastolic dysfunction or perhaps a reduction in sudden cardiac death or recurrent infarction, added Dr. Gersch, professor of medicine at the Mayo Clinic, Rochester, Minn.
Fellow discussant Dr. Steven Bolling said he agreed that STICH is a landmark trial and called the difference in outcomes between the intention-to-treat and actual treatment analyses "interesting."
Yet, "if the biological effect that our patients feel is really what treatment they receive, then under that analysis, of course, as a surgeon, you must conclude that patients with left-ventricular dysfunction should receive coronary artery bypass," added professor of surgery and director of the mitral valve clinic at the University of Michigan, Ann Arbor.
Patients in STICH were randomized at 99 medical centers in 22 countries and had a left ventricular ejection fraction of 35% or less and coronary artery disease suitable for CABG. The median time to CABG was 10 days. In all, 91% of patients received at least one arterial conduit, 86% received at least one venous conduit, and 88% received a total of at least two grafts. The median hospital stay was 9 days (range 7-13).