From Wiki Journal Club
Jump to navigation Jump to search
Reardon MJ, et al. "Trial of everolimus-eluting stents or bypass surgery for coronary disease". The New England Journal of Medicine. 2015. 372(13):1204-12.
PubMedFull textPDF

Clinical Question

In patients with multivessel coronary artery disease (CAD) undergoing revascularization, is percutaneous coronary intervention (PCI) with 2nd-generation drug-eluting stents (DES) noninferior to CABG with regard to death, MI, or target-vessel revascularization?

Bottom Line

In patients with multivessel CAD undergoing revascularization, PCI with 2nd-generation DES is inferior to CABG and is associated with a 4.7% absolute increase in death, MI, or target-vessel revascularization at 4.6 years. The primary endpoint increase with PCI was driven primarily by a 3.3% absolute increase in target-vessel revascularization and a trend towards increased nonfatal MIs.

Major Points

While the standard of care for patients presenting with symptomatic CAD attributable to multivessel (critical stenosis in ≥ 2 vessels) atherosclerosis has been surgical revascularization with CABG (also a class I indication in current consensus guidelines), this practice has been largely based on older RCT data using bare-metal stents or 1st generation DES which are known to be associated with a higher thrombosis rate than current 2nd-generation DES. Furthermore, the 2012 SYNTAX trial demonstrated that the benefit of CABG over PCI appears to depend on anatomic complexity. As a result, a contemporary comparison of PCI with CABG using contemporary 2nd-generation DES across the spectrum of anatomic complexity in multivessel CAD was needed.

The 2015 Randomized Comparison of Coronary Artery Bypass Surgery and Everolimus-Eluting Stent Implantation in the Treatment of Patients with Multivessel Coronary Artery Disease (BEST) trial randomized 880 patients with multivessel CAD to either PCI with 2nd-generation DES or CABG and assessed for a primary composite outcome of death, MI, or target-vessel revascularization. At median 4.6 years, there was a 4.7% absolute increase in the primary outcome with PCI driven primarily by 3.3% absolute increase in target-vessel revascularization and 2.1% absolute increase in nonfatal MIs (trend). In a landmark analysis looking at events after 30 days post-procedure, PCI was associated with a 2.8% absolute increase in spontaneous MIs versus CABG. In keeping with previous data suggesting superiority of CABG in diabetics, the treatment effect of CABG over PCI was manifest primarily in patients with known diabetes. Interestingly, there was no interaction between SYNTAX score and revascularization strategy, suggesting that the superiority of CABG is consistent across the spectrum of anatomic complexity although the study was not powered to assess this issue definitively. There was no difference between groups with regard to a composite safety outcome of death, MI, or stroke, although there was 23.1% absolute increase in TIMI major bleeding in the CABG group.

The results of BEST suggest that even in contemporary age of PCI using 2nd generation DES, CABG remains superior to PCI in reducing adverse cardiovascular outcomes, primarily due to superior long-term durability with fewer spontaneous MIs and reduced need for repeat revascularization. As a result, CABG is likely to remain standard of care for patients with multivessel CAD at acceptable surgical risk requiring revascularization.


As of May 2017, no guidelines have been published that reflect the results of this trial.


  • Prospective, open-label, randomized controlled clinical trial
  • N=880
    • PCI (n=438)
    • CABG (n=442)
  • Setting: 27 centers in southeast Asia
  • Enrollment: July 2008 to September 2013
  • Median follow-up: 4.6 years
  • Analysis: Intention-to-treat
  • Primary outcome: Death, nonfatal MI, and target-vessel revascularization


Inclusion Criteria

  • Age ≥ 18 years
  • Angiographically confirmed multivessel CAD (> 70% stenosis in at least two major epicardial vessels in two separate coronary artery territories)
  • Indication for revascularization based on symptoms of angina and/or object evidence of myocardial ischemia

Exclusion Criteria

  • Known hypersensitivity to heparin, ASA, clopidogrel, everolimus, steel or contrast media
  • NYHA III or greater CHF symptoms
  • Planned simultaneous surgical procedure unrelated to coronary revascularization
  • Previous bypass surgery
  • Prior PCI with DES within 1 year
  • Two or more chronic total occlusions in major coronary arteries
  • Acute STEMI within 72 hours prior
  • Abnormal CK, CKMB, or troponin levels at the time of randomization
  • Previous stroke within 6 months or stroke at more than 6 months with significant residuum
  • Life expectancy < 2 years
  • Prior history of significant bleeding within 6 months
  • Left main disease with ≥ 50% stenosis

Baseline Characteristics

From the PCI group

  • Demographics: age 64.0, male 69.4%, BMI 24.7
  • Co-morbidities: DM 40.4%, HTN 67.6%, HLD 54.6%, smoker 20.1%, CVA 8.4%, CKD 2.1%, PAD 3.4%, COPD 1.8%
  • Cardiac: Previous PCI 6.8%, previous MI 5.7%, CHF 3.7%, EF 59.1%
  • Coronary Disease: Stable angina 47.9%, unstable angina 42.2%, acute MI 9.8%, three-vessel disease 75.3%, two-vessel disease 24.7%, chronic total occlusion 28.8%, bifurcation lesion 57.5%, heavily calcified lesion 32.2%, EuroSCORE 2.9, SYNTAX 24.2


  • Randomized 1:1 to PCI or CABG
    • Randomization performed in random block sizes of 6 and 8 with stratification by center
  • Enrollment occurred after coronary angiography was performed
  • Use of intravascular ultrasound, adjunctive devices, or glycoprotein IIa/IIIb inhibitors at operator's discretion
  • All patients undergoing PCI were prescribed aspirin and clopidogrel before or during the procedure
  • After PCI all patients received aspirin 100MG daily and clopidogrel 75MG daily for at least 12 months
  • For all patients, the use of secondary-prevention medications according to clinical guidelines was strongly recommended
  • Routine angiographic follow-up strongly discouraged for all patients in the absence of signs or symptoms of ischemia
  • Immediate follow-up assessments were performed at each hospital, and follow-up assessments were performed at 30 days, 6, 9, and 12 months, and annually thereafter
  • Patients in the PCI group received an average of 3.4 stents per patient


Comparisons are PCI vs. CABG

Primary Outcomes

Death, nonfatal MI, or target-vessel revascularization
67 (15.3%) vs. 47 (10.6%) [HR 1.57, 95% CI 1.01-2.13, p=0.04]

Secondary Outcomes

All-cause death
29 (6.6%) vs. 22 (5.0%) [HR 1.34, 95% CI 0.77-2.34, p=0.30]
Myocardial infarction
21 (4.8%) vs. 12 (2.7%) [HR 1.76, 95% CI 0.87-3.58, p=0.11]
Spontaneous myocardial infarction
19 (4.3%) vs. 7 (1.6%) [HR 2.75, 95% CI 1.16-6.54, p=0.02]
11 (2.5%) vs. 13 (2.9%) [HR 0.86, 95% CI 0.39-1.93, p=0.72]
Target-vessel revascularization
31 (7.1%) vs. 17 (3.8%) [HR 1.88, 95% CI 1.04-3.40, p=0.03]

Subgroup Analysis

Diabetes (interaction p=0.06)
Yes - 19.2% vs. 9.1% (p=0.007)
No - 12.6% vs. 11.7% (p=0.79)

Adverse Events

TIMI major bleeding
30 (6.8%) vs. 132 (29.9%) [HR 0.20, 95% CI 0.14-0.30, p<0.001]


  • Results may have been driven by benefit of CABG in diabetic patients (40% of this population); these results may have no benefit in non-diabetic patients (and no such subgroup analysis performed)
  • Open-label nature of the study allows for ascertainment bias (e.g., patients randomized to PCI possibly were more likely to undergo routine coronary angiography resulting in detection of an asymptomatic lesion resulting in target-vessel revascularization).
  • Study terminated early due to slow enrollment. As a result, power is limited to detect differences between PCI and CABG in secondary outcomes and/or subgroup analyses.
  • Patients in the PCI group were more likely to receive certain medications including antiplatelets, beta blockers, ACE inhibitors or ARBs, and CCBs, potentially resulting in underestimation of the benefit of CABG over PCI.
  • Study patients geographically limited to southeast Asia, limiting generalizability of findings to other populations.


  • Study supported by CardioVascular Research Foundation and Abbott Vascular

Further Reading