Read the full EBM Focus and earn CME credit.
Reference: NOBLE trial (Lancet 2016 Oct 31 early online) (level 2 [mid-level] evidence), EXCEL trial (N Engl J Med 2016 Oct 31 early online) (level 1 [likely reliable] evidence)
- Coronary artery bypass grafting (CABG) is recommended to improve survival in most patients with left main coronary artery stenosis of ≥ 50% diameter and percutaneous coronary intervention (PCI) is considered to be a reasonable alternative in selected patients.
- In patients with unprotected left main coronary disease, PCI with a drug-eluting stent was compared to CABG in 1,201 patients in the NOBLE trial and 1,905 patients in the EXCEL trial.
- The NOBLE trial found PCI to be associated with an increased risk of the composite outcome of all-cause death, nonprocedural myocardial infarction, stroke, and any repeat coronary revascularization compared to CABG, while the EXCEL trial found a similar risk between the two procedures for the composite outcome of all-cause death, myocardial infarction, and stroke.
- The different results may be due to differences in the primary outcome evaluated and the type of stents used which were different in terms of composition and drug-eluting polymers (biodegradable vs. durable).
Left main coronary artery disease (CAD) in patients having coronary angiography has an estimated prevalence of 3.6% (Catheter Cardiovasc Interv 2006 Sep;68(3):357) and is associated with a poor prognosis in medically treated patients with severe stenosis (Circulation 2011 Dec 6;124(23):e574). The SYNTAX trial investigated the comparative efficacy of PCI with a paclitaxel-eluting stent vs. CABG in patients with left main CAD or 3-vessel disease (Lancet 2013 Feb 23;381(9867):629). In an analysis of 705 patients with left main CAD (Circulation 2014 Jun 10;129(23):2388), no difference was found in the composite outcome of all cause death, myocardial infarction, stroke, and repeat revascularization events at 5 years. In order to compare the efficacy of these two procedures for treating left main disease in a larger number of patients and with the newer generation of drug-eluting stents, 1,201 patients were evaluated in the NOBLE trial and 1,905 patients in the EXCEL trial with a median follow-up of 3 years for both trials. In the NOBLE trial, use of a biodegradable polymer biolimus-eluting stent (Biomatrix Flex) was recommended after the first 73 patients (11% received a first generation stent), and in the EXCEL trial, PCI was performed with fluoropolymer-based cobalt-chromium everolimus-eluting stents (XIENCE). Due to low event rates in the NOBLE trial, the original full 2-year follow-up was amended to a median follow-up of 3 years including all events occurring up to 5 years. Follow-up data at 5 years was available for 36% of patients.
In the NOBLE trial, PCI was associated with a higher 5-year estimated rate of the composite outcome of all-cause mortality, nonprocedural myocardial infarction, any repeat coronary revascularization, and stroke which was 29% vs. 19% with CABG (p = 0.007, NNH 10). The individual components of this composite with significantly higher rates in the PCI group were nonprocedural myocardial infarction which occurred in 7% vs. 2% (p = 0.004, NNH 20) and any repeat coronary revascularization in 16% vs. 10% (p = 0.032, NNH 16). There were no significant differences in all-cause mortality or stroke when evaluated alone.
In the EXCEL trial, PCI was found to be noninferior to CABG for the primary composite outcome of all-cause death, myocardial infarction, or stroke at 3 years because the upper limit of the 97.5% CI for the difference between the two interventions was < 4.2% higher. Noninferiority was also met for the secondary outcome which additionally included ischemia-driven revascularization. There were no significant differences in the single components of the secondary composite outcome at 3 years except for an increased rate of ischemia-driven revascularization associated with PCI which occurred in 12.6% vs. 7.5% with CABG (p < 0.001, NNH 19). At 30 days, PCI was associated with a decreased risk of both the primary composite outcome (4.9% vs. 7.9% with CABG, p = 0.008, NNT 34) and myocardial infarction (3.9% vs. 6.2% with CABG, p = 0.02, NNT 44).
The results of these trials differed for overall risk associated with PCI. The NOBLE trial found that PCI had a higher risk of overall adverse events compared to CABG while no significant difference in overall risk was found in the EXCEL trial. Some factors which differed between the trials may help explain these results. Firstly, the primary outcome in the NOBLE trial excluded procedural myocardial infarction while the EXCEL trial included it. The exclusion of these early myocardial infarctions in the primary outcome of the NOBLE trial (with a nonsignificant higher rate in the CABG group in an analysis of only 45% of the patients) may help explain the differences observed. Secondly, the trial populations differed in that a greater proportion of patients in the NOBLE trial had less complex lesions with about 91% having SYNTAX scores < 33 vs. 76% in the EXCEL trial, although if anything, the less complex lesions should have led to better PCI outcomes in the NOBLE trial which was not observed. Finally, stents with different compositions and drug-eluting polymers were used in the two trials. The NOBLE trial included the use of first generation stents in 11% of the patients or a biolimus-eluting stent with a biodegradable polymer while fluoropolymer-based cobalt-chromium everolimus-eluting stents were used in the EXCEL trial. Overall the current results of these trials suggest that the choice of procedure for patients with left main disease will have to be made on an individualized basis after balancing the potential risk difference in revascularization favoring CABG with the potential risk difference in short-term adverse effects possibly favoring PCI.
For more information, see the Left main coronary artery disease topic or Revascularization for coronary artery disease (CAD) topics in DynaMed Plus. DynaMed users click here.