Future Revascularization Evaluation in Patients with Diabetes Mellitus: Optimal Management of Multivessel Disease - FREEDOM


Current American College of Cardiology/American Heart Association guidelines recommend coronary artery bypass grafting (CABG) as the primary revascularization strategy in patients with diabetes mellitus (DM) and multivessel disease (MVD). The current trial sought to compare outcomes between DM patients with MVD who underwent percutaneous coronary intervention (PCI) versus CABG using contemporary techniques.


This trial sought to determine the optimal revascularization strategy (PCI vs. CABG) in patients with DM and MVD.

Study Design

  • Randomized
  • Blinded
  • Parallel

Patient Populations:

  • DM (type 1 or type 2) defined according to the American Diabetes Association as history of either:
    - Presence of classic symptoms of DM with unequivocal elevation of plasma glucose (2-hour post-prandial or random of >200 mg/dl), or
    - Fasting plasma glucose elevation on more than one occasion of at least 126 mg/dl
  • Angiographically confirmed multivessel coronary artery disease and amenable to either PCI or CABG
    - Critical (>70%) lesions in at least two major epicardial vessels and in at least two separate coronary artery territories (left anterior descending, left circumflex, right coronary artery)
    - Indication for revascularization based on symptoms of angina and/or objective evidence of myocardial ischemia

    Number of screened applicants: 32,966
    Number of enrollees: 1,900
    Duration of follow-up: 5 years
    Mean patient age: 63.2 years
    Percentage female: 33%
    Ejection fraction: 66%


  • Severe congestive heart failure (class III or IV)
  • Simultaneous surgical procedure
  • Prior CABG or PCI with stent within 6 months
  • Prior cardiac valve surgery
  • 2+ chronic total occlusions in major territories
  • Acute ST-elevation MI (Q-wave) within 72 hours
  • Creatine kinase (CK) >2x normal and/or abnormal CK-MB levels
  • Stroke within 6 months or >6 months with residual deficit
  • Concurrent enrollment in another clinical trial

Primary Endpoints:

  • Death/MI/stroke

Secondary Endpoints:

  • Cardiovascular death
  • TLR at 1 year
  • Major adverse cardiac events/major adverse cardiac and cerebrovascular events at 30 days and 12 months

Drug/Procedures Used:

All qualifying angiograms were reviewed by an interventionalist and cardiac surgeon. PCI was performed with drug-eluting stents (DES), either paclitaxel-eluting stents (PES) or sirolimus-eluting stents (SES). For CABG, arterial revascularization was encouraged; the left internal mammary artery was utilized in 94% of patients. The choice of surgical approach—cardiopulmonary bypass and cardioplegic arrest, conventional CABG, or “beating heart” off-pump CABG—was left to the individual surgeon’s judgment.

Concomitant Medications:

For PCI: All patients received aspirin 325 mg and ≥300 mg clopidogrel; dual antiplatelet therapy was continued for a minimum of 12 months. Abciximab was preferentially used as the procedural anticoagulant of choice, but unfractionated heparin or bivalirudin could be used at the operator’s preference. At 5 years: aspirin (93%), thienopyridine (42% in PCI, 16% in CABG), statin (90%), angiotensin-converting enzyme inhibitor (64%), angiotensin-receptor blocker (37%), beta-blockers (79%).

Principal Findings:

A total of 1,900 patients were randomized, 953 to PCI and 947 to CABG. Baseline characteristics were fairly similar between the two arms. The mean baseline glycated hemoglobin was 7.8%, and 32% were on insulin. Approximately 16% were current smokers, with previous myocardial infarction (MI) in 26%, stable angina in 69%, and recent acute coronary syndrome in 31%. The mean ejection fraction (EF) was 66%, and only 3% had evidence of left ventricular (LV) EF <40%. About 84% had three-vessel disease, the mean number of lesions was 5.7, and the mean SYNTAX score was 26.2 (35% low, 45% moderate, 20% high). In the PCI arm, staged PCI was performed in 34%, with a mean of 3.5 lesions stented per patient and a mean stented length of 26.1 mm. CABG was performed off-pump in 19%, and the mean number of graft vessels was 2.9.

The primary composite outcome of death, MI, and stroke at 5 years was significantly higher in the PCI arm as compared with the CABG arm (26.6% vs. 18.7%, p = 0.005), driven by significant reductions in all-cause mortality (16.3% vs. 10.9%, p = 0.049) and MI (13.9% vs. 6.0%, p < 0.001); strokes were lower in the PCI arm (2.4% vs. 5.2%, p = 0.03). Thirty-day events for the primary endpoint were lower in the PCI arm (26 vs. 42 events); however, starting at 2 years (13.0% vs. 11.9%), the curves continued to diverge out on longer-term follow-up. Results were robust across numerous subgroups, including baseline SYNTAX score (23.2% vs. 17.2% in the low SYNTAX group). At 1 year, repeat revascularization was significantly higher in the PCI arm (13% vs. 5%, p < 0.0001). Thirty-day bleeding rates were similar (p = 0.13).

All events were higher in patients with insulin-dependent DM as compared with non–insulin-dependent DM, including the primary endpoint (29% vs. 19% at 5 years, p < 0.001). However, the benefit of CABG over PCI was sustained for all endpoints studied.

Cost-effectiveness: Total procedural cost for PCI vs. CABG was $9,739 vs. $13,014; index hospitalization costs were $25,845 vs. $34,467; difference = $8,622 (p < 0.001). However, over 5 years, the difference in costs was $3,641. CABG was highly cost-effective (incremental cost-effectiveness ratio [ICER]: $8,132/quality-adjusted life-year [QALY] gained). This was similar across all SYNTAX categories. Even under the highly conservative assumption of no further benefit of CABG on either survival or costs beyond 5 years, the ICER remained relatively favorable at $27,022/QALY gained.

Quality-of-life assessments: This was assessed using the Seattle Angina Questionnaire in three domains: angina frequency, physical limitations, and quality of life. Baseline values were similar between PCI and CABG patients, except for a mild but statistical difference between PCI and CABG patients for physical limitations (69.9 vs. 67.3). Angina frequency was slightly better in CABG patients than PCI at 2 years (mean difference 1.3). Scores for the other two domains were better for PCI at 1 month, but then better for CABG over PCI between years 1 and 3. Beyond 3 years, no differences were noted in the three domains of quality of life between the two strategies.


The results of the landmark FREEDOM trial indicate that in diabetic patients with multivessel disease, CABG is superior to DES PCI, and should remain the revascularization strategy of choice in this patient population. CABG resulted in lower rates of death and MI, but higher risk of stroke. This is true for patients with either insulin-dependent or non–insulin-dependent DM.

Although initial hospitalization costs were higher with CABG, it is highly cost-effective on long-term follow-up, with cost-effectiveness thresholds well below currently accepted norms in the United States ($50,000/QALY). Quality-of-life analyses indicate a modest clinical benefit with CABG over PCI in the intermediate-term (up to 2 years), but no differences over long-term follow-up.

A number of earlier studies have studied this question in this patient population. The BARI trial reported superior outcomes with CABG over balloon angioplasty in the 1990s. More recently, the CARDia trial and a subgroup analysis of the SYNTAX trial demonstrated superior outcomes with CABG over DES PCI, mainly due to reductions in repeat revascularization. However, they were underpowered for individual clinical outcomes. In this trial, CABG was noted to result in superior outcomes not only due to reductions in target lesion revascularization (TLR), but also due to reductions in MI and all-cause mortality.

Although these trials utilized first-generation DES (SES/PES), results are unlikely to be significantly better with newer-generation stents (everolimus-eluting stents or zotarolimus-eluting stents) since their benefit over first-generation DES is mainly in reductions in TLR. Longer-term follow-up of the FREEDOM trial is awaited.


Presented by Dr. George Dangas at the Transcatheter Cardiovascular Therapeutics meeting (TCT 2013), San Francisco, CA, October 31, 2013.

Farkouh ME, Domanski M, Sleeper LA, et al., on behalf of the FREEDOM Trial Investigators. Strategies for multivessel revascularization in patients with diabetes. N Engl J Med 2012;367:2375-84.

Magnuson EA, Farkouh ME, Fuster V, et al., on behalf of the FREEDOM Trial Investigators. Cost-effectiveness of percutaneous coronary intervention with drug eluting stents versus bypass surgery for patients with diabetes mellitus and multivessel coronary artery disease: results from the FREEDOM trial. Circulation 2013;127:820-31.

 Abdallah MS, Wang K, Magnuson EA, et al., on behalf of the FREEDOM Trial Investigators. Quality of life after PCI vs CABG among patients with diabetes and multivessel coronary artery disease: a randomized clinical trial. JAMA 2013;310:1581-90

Presented by Dr. Valentine Fuster at the American Heart Association Scientific Sessions, San Francisco, CA, November 4, 2012.

FREEDOM Cost-Effectiveness: Presented by Dr. Elizabeth Magnuson at the American Heart Association Scientific Sessions, Los Angeles, CA, November 4, 2012.

Clinical Topics: Acute Coronary Syndromes, Cardiac Surgery, Invasive Cardiovascular Angiography and Intervention, Prevention, Stable Ischemic Heart Disease, Atherosclerotic Disease (CAD/PAD), Aortic Surgery, Cardiac Surgery and SIHD, Interventions and ACS, Interventions and Coronary Artery Disease, Diet, Chronic Angina

Keywords: Coronary Artery Disease, Myocardial Infarction, Stroke, Acute Coronary Syndrome, Angina, Stable, Drug-Eluting Stents, Insulins, Sirolimus, Angioplasty, Balloon, Coronary, Glucose, Hemoglobin A, Glycosylated, Paclitaxel, Quality of Life, Mammary Arteries, Cardiopulmonary Bypass, Coronary Artery Bypass, Hospitalization, Diabetes Mellitus, Fasting

< Back to Listings