Treatments for Coronary Artery Disease and Left Ventricular Dysfunction

New Orleans—Combining coronary artery bypass grafting (CABG) with medical therapies for patients with coronary artery disease (CAD) and heart failure increased the overall risk of death for approximately 2 years after surgery compared with medication alone; however, in subsequent years, the rate of all-cause mortality in the CABG group dropped below the rate seen in the medication-only group, according to researchers with the STICH (Surgical Treatment for Ischemic Heart Failure) trial. Data were presented at the ACC meeting by STICH investigator Eric J. Velazquez, MD, associate professor of medicine, division of cardiovascular medicine, Duke University Medical Center, Clinical Research Institute, Durham, North Carolina, and published in the New England Journal of Medicine [2011;364(17):1607-1616]. Although the difference between the treatment arms was not significant on the primary end point of all-cause mortality at 5 years, the CABG arm demonstrated significantly lower rates of cardiovascular mortality and of death from all causes combined with hospitalization for cardiovascular reasons at 5 years, which were secondary end points of the study. CAD is the most frequent precursor to heart failure, an often fatal event that strikes nearly 21 million people in the United States and Europe annually, and a major cause of left ventricular dysfunction (LVD). Current practices regarding CABG use in patients with CAD have been informed primarily by studies that excluded patients with severe LVD and predate several noteworthy advances in cardiac care. To clarify the role of CABG in the care of patients with concomitant CAD and LVD, STICH investigators enrolled 1212 patients from July 2002 to May 2007 for a multinational, randomized, controlled trial and followed them for a median of 56 months. STICH required that, in addition to having CAD, participants demonstrate a left ventricular ejection fraction (LVEF) ≤0.35 within 3 months of enrollment and be considered suitable candidates for CABG and medical therapy. Anyone experiencing acute myocardial infarction in the 30 days prior to enrollment or cardiogenic shock within 72 hours ofrandomization, or with aortic valve disease requiring repair, or who had >1 CABG surgery, was excluded. Patients were randomized to medication alone (n=602) or medication plus CAGB (n=610), and patient characteristics were well balanced between the study arms. In each group, 88% of patients were men and approximately one third were black. Slightly more than three quarters of patients in each arm had a history of myocardial infarction, approximately 40% had a diabetes diagnosis, and 91% had multivessel disease. Median LVEF in the medication-only arm was 28% versus 27% in the CABG group; median ages for the 2 groups were 59 years and 60 years, respectively. After randomization, a cardiologist at each participating center evaluated patients to determine the best treatment plan as suggested by current guidelines on CAD and heart failure. Commonly prescribed medications in both groups included aspirin, warfarin, beta-blockers, statins, angiotensin-converting enzyme inhibitors, and angiotensin II receptor blockers. The trial intended for an experienced surgeon to perform CABG within 14 days of randomization. A total of 555 patients in the CABG group underwent surgery as planned; most procedures were considered elective, but 5% (n=26) were performed with urgency. For 63 (11%) patients, a mitral valve operation was performed concurrently with CABG. Vast majorities received ≥1 arterial conduit and ≥1 venous conduit (91% and 86%, respectively), and 87% had ≥2 distal anastomoses placed. CABG patients spent a median of 9 days in the hospital, with a median of 52.4 hours in the intensive care unit. The primary analysis included a median of 56 months of follow-up data for 1207 (99.6%) of the 1212 STICH participants. During this time, 244 (41%) patients in the medication-only arm and 218 (36%) in the CABG group died (hazard ratio [HR] with CABG, 0.86; 95% confidence interval [CI], 0.72-1.04; P=.12). Looking only at cardiovascular-related mortality, investigators reported 201 (33%) deaths in the medication-only group compared with 168 (28%) in the CABG arm, a difference that was statistically significant (HR with CABG, 0.81; 95% CI, 0.66-1.00; P=.05). The groups also reflected a statistically significant difference between the proportion of patients in each who died or were hospitalized for cardiovascular causes, which encompassed 411 (68%) patients in the medication-only arm and 351 (58%) patients in the CABG group (HR with CABG, 0.74; 95% CI, 0.64-0.85; P<.001). According to the authors, all other end points favored CABG with the exception of 30-day mortality; they had expected this would be higher in the CABG group because of the inherent risks of surgical intervention. During the study, both arms experienced crossover; Dr. Velazquez said 55 patients assigned to CABG received medication only, whereas 65 patients randomized to medication alone subsequently underwent CABG in the first year after randomization, most often for progressive symptoms. In contrast to the intent-to-treat analysis, a 1-year as-treated analysis did find a significant difference between the groups in the rate of all-cause mortality favoring CABG (HR with CABG, 0.70; 95% CI, 0.58-0.84; P<.001). Authors hypothesized that crossover, as well as the study’s limited power and duration of follow-up, diluted the perceived effectiveness of CABG and contributed to the study’s failure to meet its primary end point. Various subgroup analyses were unable to identify any patient characteristics likely to have had a significant effect on the primary outcome. The study authors cautioned that because the primary end point failed to achieve significance, analyses of secondary outcomes demonstrating significant benefit with CABG should be considered provisional. For obvious reasons, the trial could not be blinded, and the authors cited this as one of the study’s major limitations. In Dr. Velazquez’s presentation, he recommended CAD be assessed “among all patients presenting with heart failure.” He noted that the intent-to-treat analysis showed a 19% reduction in the relative risk of cardiovascular death with CABG plus medication versus medication only. In addition, the relative risk reduction on the combined end point of all-cause mortality and cardiovascular hospitalization was 24% with CABG. As a result, Dr. Velazquez recommended CABG be considered for all heart failure patients receiving medical therapy for CAD, as a means of reducing the risk of cardiovascular mortality and morbidity. The STICH study plans to follow the patients for an additional 5 years, investigating the durability of CABG benefits in the ongoing STICH extension study.