BACKGROUND: Coronary angioplasty and coronary artery bypass grafting (CABG) are both major techniques for the management of coronary artery disease, but CABG is associated with a lower incidence of repeat revascularization. Recent studies comparing angioplasty with stenting vs CABG have yielded conflicting results, with some suggesting improved survival with stenting, and others the opposite. We thus undertook a systematic overview of the randomized trials comparing stenting vs CABG in coronary artery disease. METHODS: MEDLINE (January 1986-February 2003), ISI Current Contents, the Cochrane Controlled Trial Register, LILACS and the American Heart Association, American College of Cardiology, European Society of Cardiology, and Transcatheter Cardiovascular Therapeutics conference proceedings were among the databases we searched. Abstraction was performed in a non-blinded manner on pre-specified forms. The random-effect odds ratios for death, myocardial infarction, stroke, repeat revascularization, and symptomatic angina were computed for the longest available follow-up. RESULTS: Nine randomized trials (3283 patients, representing only 6% of all screened subjects) with an average follow-up of 28 months were included in the analysis, while four studies were excluded because they were still unpublished, ongoing, or with non-systematic stenting. No study used drug-eluting stents. The odds ratios for stenting vs CABG were 0.82 (95% confidence interval-CI 0.57-1.18, p = 0.3) for the occurrence of death, non-fatal myocardial infarction or stroke, 4.6 (95% CI 3.5-5.9, p < 0.00001) for repeat revascularization, and 2.3 (95% CI 1.8-2.8, p < 0.00001) for symptomatic angina. Heterogeneity tests were not statistically significant. The results of sensitivity analysis were similar even after stratification for single vessel, off-pump, single center or high-quality studies. CONCLUSIONS: Overall and event-free survival after conventional stenting for coronary artery disease are similar to those after CABG, but surgery is still associated with a significantly lower incidence of repeat revascularization and symptoms. The role of next-generation drug-eluting stents in widening the indications for stenting and overcoming restenosis will need to be assessed in future observational and randomized studies comparing stenting vs CABG.
BACKGROUND: Coronary angioplasty and coronary artery bypass grafting (CABG) are both major techniques for the management of coronary artery disease, but CABG is associated with a lower incidence of repeat revascularization. Recent studies comparing angioplasty with stenting vs CABG have yielded conflicting results, with some suggesting improved survival with stenting, and others the opposite. We thus undertook a systematic overview of the randomized trials comparing stenting vs CABG in coronary artery disease. METHODS: MEDLINE (January 1986-February 2003), ISI Current Contents, the Cochrane Controlled Trial Register, LILACS and the American Heart Association, American College of Cardiology, European Society of Cardiology, and Transcatheter Cardiovascular Therapeutics conference proceedings were among the databases we searched. Abstraction was performed in a non-blinded manner on pre-specified forms. The random-effect odds ratios for death, myocardial infarction, stroke, repeat revascularization, and symptomatic angina were computed for the longest available follow-up. RESULTS: Nine randomized trials (3283 patients, representing only 6% of all screened subjects) with an average follow-up of 28 months were included in the analysis, while four studies were excluded because they were still unpublished, ongoing, or with non-systematic stenting. No study used drug-eluting stents. The odds ratios for stenting vs CABG were 0.82 (95% confidence interval-CI 0.57-1.18, p = 0.3) for the occurrence of death, non-fatal myocardial infarction or stroke, 4.6 (95% CI 3.5-5.9, p < 0.00001) for repeat revascularization, and 2.3 (95% CI 1.8-2.8, p < 0.00001) for symptomatic angina. Heterogeneity tests were not statistically significant. The results of sensitivity analysis were similar even after stratification for single vessel, off-pump, single center or high-quality studies. CONCLUSIONS: Overall and event-free survival after conventional stenting for coronary artery disease are similar to those after CABG, but surgery is still associated with a significantly lower incidence of repeat revascularization and symptoms. The role of next-generation drug-eluting stents in widening the indications for stenting and overcoming restenosis will need to be assessed in future observational and randomized studies comparing stenting vs CABG.