Log in to MyACC
PCI vs. CABG for Left Main Disease
Study Questions:
What is the optimal method of revascularization for left main coronary artery disease (LMCAD)?
Methods:
This is a meta-analysis of prospective randomized clinical trials investigating percutaneous coronary intervention (PCI) with drug-eluting stents (DES) versus coronary artery bypass grafting (CABG) for the treatment of LMCAD. The primary endpoint was all-cause mortality. Secondary endpoints were cardiac/cardiovascular death, stroke, myocardial infarction (MI), and unplanned revascularization. Outcomes were analyzed as intention-to-treat. Heterogeneity was assessed with the I2 statistic. Sensitivity analyses were performed excluding the trial with only 1-year follow-up and individually excluding each trial from the analysis.
Results:
Five trials were included in the meta-analysis: Boudriot et al. (2011), PRECOMBAT, NOBLE, EXCEL, and SYNTAX. From the five trials, there were 4,612 patients, of which 2,303 had been randomized to PCI with DES and 2,309 to CABG. Follow-up was 1 year for the Boudriot et al. trial; 5 years for PRECOMBAT, NOBLE, and EXCEL; and 10 years for SYNTAX. Weighted mean follow-up was 67.1 months. At last follow-up, there was no significant difference in all-cause mortality between PCI with DES versus CABG (relative risk [RR], 1.03; 95% confidence interval [CI], 0.82-1.30; p = 0.779) with moderate heterogeneity (I2 = 42.9%), and cardiovascular death rates were similar (RR, 1.03; 95% CI, 0.79-1.34; p = 0.817; I2 = 0.0%). Between PCI with DES and CABG, there was no significant difference in stroke (RR, 0.74; 95% CI, 0.36-1.50; p = 0.400) or all MI (RR, 1.22; 95% CI, 0.96-1.56; p = 0.110). PCI with DES was associated with higher unplanned revascularization compared to CABG (RR, 1.73; 95% CI, 1.49-2.02; p < 0.0001). These results held with sensitivity analyses.
Conclusions:
Meta-analysis of all randomized clinical trials performed investigating the treatment of LMCAD with PCI with DES versus CABG showed no significant difference in all-cause mortality, cardiac death, stroke, or MI with follow-up to 5-6 years. Unplanned revascularization was less common after CABG.
Perspective:
The limitations of this meta-analysis are well-addressed by the authors in the Discussion section. Specifically, the authors point out differences in trial endpoints, use of different stent platforms over time, and the small number of trials ultimately included in the analysis. Follow-up also differed among the trials, and the majority of patients included had follow-up out to only 5 years. Longer-term follow-up may be needed to see if there is a difference between PCI and CABG for the treatment of LMCAD. The unfortunate difference in definitions of trial endpoints, common to many meta-analyses, speaks to the need for what the authors refer to as “harmonization,” a good example of which would be that performed by the Valve Academic Research Consortium. This would allow for better comparison between and pooling of data from multiple trials.
As has also been pointed out before, trials comparing treatments with low incidence of mortality have led to design of trials with composite endpoints in order to have enough events to determine statistical significance. The inclusion of “soft” endpoints such as revascularization in coronary disease, or readmission for heart failure in transcatheter aortic valve replacement, can dilute the analysis and muddy the final conclusion.
Clinical Topics: Arrhythmias and Clinical EP, Cardiac Surgery, Invasive Cardiovascular Angiography and Intervention, Atherosclerotic Disease (CAD/PAD), SCD/Ventricular Arrhythmias, Aortic Surgery, Cardiac Surgery and Arrhythmias, Interventions and Coronary Artery Disease
Keywords: Cardiac Surgical Procedures, Coronary Artery Bypass, Coronary Artery Disease, Death, Sudden, Cardiac, Drug-Eluting Stents, Myocardial Infarction, Myocardial Revascularization, Percutaneous Coronary Intervention, Risk, Stroke
< Back to Listings
