What patterns of coronary disease are associated with improved short and long term survival after CABG compared with medical therapy?

Many factors have influenced the short and long term results of bypass surgery, not least the improvements in surgical techniques and experience, changes in the population of patients undergoing surgery, many of whom would never have been deemed suitable for surgery even 10 years ago, improvements in postoperative medical management and the use of the left internal mammary artery (LIMA) as the graft of choice for the left anterior descending coronary artery (LAD) in virtually all patients today.

30 day operative mortality
Short term survival after bypass surgery is 1–3% at most institutions around the world. The Society of Thoracic Surgeons National Database mortality figures1 for 80,881 patients undergoing isolated bypass surgery between 1980 and 1990 were 4.75% for left main disease, 3.32% for triple vessel disease and 2.86% for one and two vessel disease. In-hospital mortality was 2.9% for first time operation and 7.14% for re-operation. Recognised factors affecting in-hospital mortality include older age, female sex, co-morbid renal and cardiovascular disease, diabetes, cardiogenic shock, emergency, salvage or redo operation, preoperative intra-aortic balloon pump use and associated valve disease.

Long term survival after surgery
The late results of bypass surgery depend on the extent of cardiac disease, the effectiveness of the original operation, progression rate of atherosclerosis and the impact of non-cardiac disease. Patient-related variables associated with poorer late survival include reduced ventricular function, congestive cardiac failure, triple vessel or left main stem disease, severity of symptoms, advanced age and diabetes.

The patients who gain most from surgery are those most at risk from dying with medical therapy alone. Pertinent high-risk characteristics included left main stem (LMS) disease, triple vessel disease or double vessel disease that included a proximal LAD lesion, and triple vessel disease associated with impaired LV function. The VA study at 18 years2 demonstrated superior surgical survival throughout the 18 years, but was only significant overall at 7 years (med. vs surg. survival 53% vs 79% p = 0.007); benefit was much greater in the high risk group with LMS stenosis >50%, single or double vessel disease with impaired LV function, and triple vessel disease with LV EF <40%. p =" 0.0002)" p =" 0.01)">95% in two or three vessel disease was an outstanding anatomical predictor of survival (med. vs surg. survival at 10 years 65% vs 77% (p = 0.007)), again with significant crossover into the surgery group. The CASS study4 demonstrated no difference in survival for any subset at 5 years, but did not include any patients with poor LV function, LMS disease, angina greater than class 2, co-morbid disease or unstable angina. It is therefore difficult to extrapolate data from this trial to modern patient populations.

Combining results from seven of these early randomised trials led to the publication of survival figures for 5, 7 and 10 years. Medical vs surgical mortality for all patients was 15.8% vs 10.2% (p = 0.0001) at 5 years, with attenuation of this benefit to a mortality of 30.5% (med.) and 26.4% (surg.) (p = 0.03) at 10 years. Extension of life for all patients having surgery was 4.3 months at 10 years. High-risk patients once again benefited the most from surgery, but in lower risk groups, a survival extension for those with proximal LAD disease (14 months), triple vessel disease (7 months) or LMS disease (19 months) was identified.

This survival benefit was independent of degree of LV impairment or abnormal stress testing. Median survival for patients with LMS disease was 13.1 years in the surgical group and 6.2 years for those treated medically. The superior patency of the LIMA graft compared with saphenous vein grafts has been established beyond any doubt and additional survival benefit, up to 18 years, has been demonstrated.