Single versus bilateral internal mammary artery grafts: 10-year outcome analysis

The superior long-term patency of the internal mammary artery (IMA) confers important short-term and late survival advantages when grafted to the left anterior descending coronary artery. However, it remains uncertain whether patients derive additional survival benefit when both IMAs are used in coronary revascularization.

Between June 1983 and May 1986, 160 patients (mean age 60 years) received bilateral IMA grafts for coronary artery bypass procedures, and in 93% of patients, the right IMA was used to bypass the left coronary system. During a similar interval, a group of 161 patients matched for symptomatic status and extent of disease (mean age, 62 years) received a single left IMA and saphenous vein grafts.

The two groups were similar with respect to gender, preoperative angina class, priority status, extent of coronary artery disease, left ventricular function, and number of distal anastomoses. Diabetes was more prevalent in the patient group receiving a single IMA graft (27% versus 17.5%; p = 0.05). Early outcome was similar in the two groups; operative mortality was 0.6% for the patient group receiving single IMA grafts and 0% for those with bilateral IMA grafts. The mean follow-up of 320 hospital survivors was 10 years. Univariate analysis revealed significantly fewer overall deaths in the patients receiving bilateral IMA grafts (n = 30; p = 0.05), and less late cardiac mortality (n = 12; p = 0.016). Ten-year actuarial survival for patients dismissed from the hospital was 76% for those receiving single IMA graft versus 85% for those receiving bilateral IMA grafts. Multivariate analysis revealed diabetes (risk ratio = 1.73), advancing age (risk ratio = 1.08), and lower ejection fraction (risk ratio = 1.01) to be the only significant predictors of late cardiac death. Use of a single IMA graft was not significant (p = 0.138) despite a risk ratio of 1.78. Use of only a single IMA graft correlated with an increased risk of angina recurrence (p < 0.001), late myocardial infarction (p = 0.019), and risk of any cardiac event (p < 0.001).

Independent risk factors for late death were diabetes mellitus, older age, and reduced ejection fraction. Patients receiving bilateral IMA grafts had better long-term survival than those with a single IMA graft, but this was not independent of diabetes. Multivariate analysis, however, did confirm that compared with single arterial grafts, bilateral IMA grafting was an independent predictor of lower rates of angina recurrence, late myocardial infarction, and the composite end point of any cardiac event.