Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: Early coronary bypass graft surgery lowers mortality

ACP J Club. 1995 Mar-April;122:29. doi:10.7326/ACPJC-1995-122-2-029

Source Citation

Yusuf S, Zucker D, Peduzzi R, et al. Effect of coronary artery bypass graft surgery on survival: overview of 10-year results from randomised trials by the Coronary Artery Bypass Graft Surgery Trialists Collaboration. Lancet. 1994 Aug 27;344:563-70.



To compare initial coronary artery bypass grafting (CABG) with initial medical therapy regarding effects on mortality at 5, 7, and 10 years in patients with stable coronary heart disease (CHD) using meta-analysis.

Data sources

{Studies were identified using MEDLINE, scanning bibliographies of relevant papers, and consulting with experts.}* All principal investigators of identified studies were invited to participate in the collaborative effort and were asked if any relevant trials were missed.

Study selection

Studies selected were randomized controlled trials of patients with stable CHD (stable angina not severe enough to necessitate surgery on grounds of symptoms alone, or myocardial infarction [MI]) who were assigned to CABG or medical treatment and followed for ≥ 10 years.

Data extraction

Data were extracted on age; sex; ejection fraction; severity of angina; history of MI, hypertension, heart failure, diabetes, and smoking; ST depression at rest and exercise; medication use; number of diseased vessels; location of disease; and mortality. Individual patient data were collected from selected trials and were transmitted for central data analysis. Kaplan-Meier curves were generated for total mortality using an intention-to-treat method at 5-, 7-, and 10-year follow-ups.

Main results

7 studies met the selection criteria, with 1324 patients assigned to CABG and 1325 to medical management. 94% of patients assigned to CABG had surgery, and 37% of patients assigned to medical treatment crossed over to surgery. The CABG group had lower mortality than the medical treatment group at 5 (P < 0.001), 7 (P < 0.001), and 10 years (P = 0.03) (Table). At 5 years, the relative risk reduction (RRR) with CABG was greater in patients with left main artery disease (68%) than in those with disease in 3 vessels (42%) or in 1 or 2 vessels (23%). The RRR was similar across other patient categories, but the absolute benefit of CABG was greatest among patients at highest risk. The average life extension by CABG was 4.3 months, but the benefit differed according to left ventricular (LV) function (10.6 mo with reduced LV function vs 2.3 mo with normal LV function) and extent of coronary artery disease (19.3 mo for left main, 5.7 mo for 3-vessel, and 1.8 mo for 1- or 2-vessel disease).


Early coronary artery bypass graft surgery reduces total mortality compared with initial medical therapy in patients with stable coronary heart disease.

Sources of funding: National, Heart, Lung, and Blood Institute and Veterans Health Service and Research Administration.

For article reprint: Dr. S. Yusuf, 252 HGH-McMaster Clinic, Hamilton General Hospital, Hamilton, Ontario L8L 2X2, Canada. FAX 905-521-1166.

*Information provided by author.

Table. Coronary artery bypass graft (CABG) surgery vs medical therapy for stable coronary heart disease (7 trials)†

Outcomes CABG Medical therapy RRR (95% CI) NNT (CI)
Mortality at 5 y 10.2% 15.8% 39% (23 to 52) 18 (13 to 30)
Mortality at 7 y 15.8% 21.7% 32% (17 to 44) 17 (12 to 31)
Mortality at 10 y 26.4% 30.5% 17% (2 to 30) 25 (14 to 153)

†Abbreviations defined in Glossary; NNT and CI calculated from data in article.


Although randomized trials have shown that CABG surgery is more effective than medical therapy in relieving anginal symptoms and reducing myocardial ischemia, the effect of CABG on survival is controversial. Most trials have shown no improvement in overall survival, but the subgroup at highest risk in each trial did show reduced mortality. None of the randomized trials was large enough, however, to have sufficient statistical power to document clinically important reductions in risk (e.g., 20% to 30%) by CABG. The investigators of the 7 randomized trials have collaborated to pool primary data comparing CABG with medical therapy for stable angina, thereby substantially improving statistical power. The finding that CABG significantly improves survival finally should lay to rest the controversy about inconsistencies among the randomized trials. The overall risk reduction by CABG of 39% at 5 years is conservative given the high crossover rate among medically treated patients.

Which patients should be offered CABG? The data show that the amount of survival prolongation varies among patients with different clinical characteristics; therefore, therapy should be individualized (1). Among patients at highest risk (e.g., left main disease, multivessel disease with reduced ejection fraction, severe myocardial ischemia), CABG can prolong survival to a clinically important degree. Although CABG may also prolong survival in low-risk patients, the degree of benefit is small enough that the choice of therapy should rest more on anticipated effects on the symptoms, quality of life, and cost-effectiveness.

Mark A. Hlatky, MD
Stanford University School of MedicineStanford, California, USA


1. Califf RM, Harrell FE Jr, Lee KL, et al. The evolution of medical and surgical therapy for coronary artery disease. A 15-year perspective. JAMA. 1989;261:2077-86.