Clinical evaluation of single versus multiple mammary artery bypass

The superior patency and clinical advantages of internal mammary artery (IMA) grafting are well established. However, the relative benefits of routine multiple IMA grafting remain uncertain. To determine whether routine multiple compared with single IMA utilization improved survival of patients undergoing coronary bypass procedures, 1,063 patients were prospectively allocated, beginning in 1984, to divergent management strategies of single (group 1, n=420) versus multiple IMA grafting (group 2, n=643). Subsequent analysis of anatomical extent of disease and preoperative baseline risk factors showed no differences (p=NS) between the two groups. All variables reflecting operative technique were similar (p=NS) for the two groups, except 74% of group 1 patients with multivessel disease received a single IMA graft, whereas 71% of group 2 patients with multivessel disease received multiple IMAs (p<0.05). By multivariate analysis, impairment of left ventricular ejection fraction, acute evolving myocardial infarction, advanced age, and unstable angina were incremental risk factors for mortality (all p<0.03), but group assignment (p=0.4) and ultimate therapy were not (p=0.6). Survival probabilities (expressed as 30-day group 1/group 2; 4-year group, 1/group 2) were overall (0.97/0.98; 0.93/0.90), elective (0.98/0.99; 0.97/0.92), acute (0.95/0.97; 0.89/0.88), age of less than 65 years (0.98/0.99; 0.97/0.93), age of 65 years or older (0.93/0.97; 0.84/0.89), ejection fraction of 0.40 or more (0.97/0.99; 0.95/0.94), ejection fraction of less than 0.40 (0.95/0.96; 0.87/0.82), nondiabetic (0.98/0.98; 0.94/0.91), and diabetic (0.92/0.97; 0.88/0.87). No differences in survival were significant (p=NS). Incidence of sternal infection (2.1% in group 1 versus 0.9% in group 2, p=NS) and reoperation for bypass grafting (0.9% in group 1 versus 0.9% in group 2, p=NS) also did not differ. These data suggest that routine multiple IMA grafting results in negligible survival benefit beyond that attributable to a single IMA, at least at intermediate-term (average, 4 years) follow-up.

Duke Scholars

Author Lawrence H. Muhlbaier Biostatistics & Bioinformatics, Division of Biostatistics