Skip to main content

Fractional Flow Reserve–Guided PCI as Compared with Coronary Bypass Surgery

What was known?

A large body of clinical trial data has shown that patients with three-vessel coronary artery disease (CAD) have better outcomes with coronary-artery bypass grafting (CABG) than with percutaneous coronary intervention (PCI). However, these trials largely involved first generation stents and did not use fractional flow reserve (FFR) to guide PCI. The use of second-generation stents and FFR has improved the outcomes of PCI.


The aim of this study was to demonstrate the non-inferiority of FFR-guided PCI with current generation drug eluting stents (DES) compared with CABG in three-vessel disease via major adverse cardiac or cerebrovascular events.

What this study adds:

This study provides a head-to-head comparison of two revascularisation strategies using contemporary PCI techniques in patients with multivessel disease, which has been lacking in the literature.

Clinical implications:

This study confirms that FFR-guided PCI with current-generation DES did not meet criteria for non-inferiority, when compared with CABG in patients with three-vessel CAD.

Study design:

The Fractional Flow Reserve versus Angiography for Multivessel Evaluation (FAME) 3 study (NCT02100722) was a multicentre, international, randomised, controlled noninferiority trial. Patients with three-vessel CAD (not involving the left main) were randomly allocated to undergo CABG (n=743) or PCR (n=757). Patients assigned to PCI underwent FFR assessment with a coronary pressure wire and intravenous or intracoronary adenosine. Only stenoses with an FFR of 0.80 or lower were eligible for treatment with PCI, which employed current generation durable polymer zotarolimus-eluting stents. Post-PCI FFR assessment was recommended. CABG was performed according to the standard practiced of the centre. Noninferiority of FFR-guided PCI to CABG was defined as an upper boundary of <1.65 for the 95% confidence interval (CI) of the hazard ratio (HR).

Primary endpoints:

  • Incidence within 1 year of a major adverse cardiac or cerebrovascular event, defined as death from any cause, myocardial infarction, stroke, or repeat revascularisation

Secondary endpoints:

  • Composite of death, myocardial infarction (MI) or stroke

Inclusion criteria:

  • Angiographically identified three-vessel CAD, defined as ≥50% diameter stenosis assessed visually in each of the three major epicardial vessels or major side branches but not involving the left main coronary artery
  • Stenosis amenable to revascularisation using either PCI or CABG, as assessed by the heart team at the trial site

Notable exclusion criteria:

  • Recent ST-segment elevation MI
  • Cardiogenic shock
  • Left ventricular ejection fraction < 30%


  • 1500 patients underwent randomisation and follow-up occurred in 99.7%
  • Patients undergoing PCI received a mean (±SD) of 3.7±1.9 stents.
  • Patients undergoing CABG received 3.4±1.0 distal anastomoses
  • At 1 year, incidence of the primary end point was 10.6% in the FFR-guided PCI treatment arm and 6.9% in the CABG arm (hazard ratio [HR], 1.5; 95% confidence interval [CI], 1.1– 2.2; P = 0.35 for noninferiority, i.e. criteria for noninferiority not met)
  • The incidence of death, MI or stroke was 7.3% in the FFR-guided PCI group and 5.2% in the CABG group (HR, 1.4; 95% CI, 0.9–2.1)
  • There were higher incidences of major bleeding, arrhythmia and acute kidney injury in the CABG group than in the FFR-guided PCI group
  • When stratified for SYNTAX score, 1-year incidence of primary endpoint was as follows: low SYNTAX (0-22): 5.5% vs. 8.6%, intermediate (23-32): 13.7% vs. 6.1%, high (>32): 12.1% vs. 6.6% for PCI vs CABG respectively


In patients with three-vessel disease, FFR-guided PCI using current generation zotarolimus-eluting stents failed to demonstrate non-inferiority to CABG. Subgroup analysis suggested that PCI was beneficial in patients with low SYNTAX scores (0-22), but this should be confirmed in future studies.


  • Relatively short duration of follow-up. 3- and 5-year follow-up is ongoing
  • Intravascular imaging was used in only 12% of patients