CABG beats PCI in Diabetics with Multivessel CAD

Abstract & Commentary

By Andrew J. Boyle, MBBS, PhD, Assistant Professor of Medicine, Interventional Cardiology, University of California, San Francisco. Dr. Boyle reports no financial relationships relevant to this field of study.

Source: Farkouh ME, et al. Strategies for multivessel revascularization in patients with diabetes. N Engl J Med 2012;367:2375-2384.

Among patients with coronary artery disease (CAD), those with diabetes are at higher risk for adverse outcomes. Thus, it is important to find the optimal treatment strategy for this high-risk group to improve their prognosis. Subgroup analyses of diabetics in prior studies of coronary artery bypass graft (CABG) surgery vs percutaneous coronary intervention (PCI) have suggested that CABG would result in superior outcomes. To test this hypothesis, Farkouh and colleagues performed the prospective, randomized, controlled Future Revascularization Evaluation in Patients with Diabetes Mellitus: Optimal Management of Multi-vessel Disease (FREEDOM) trial of CABG vs PCI in diabetics with multivessel CAD.

This was a multicenter study of 1900 patients in 140 centers around the world. The inclusion criteria were diabetes and angiographically confirmed multivessel CAD with stenosis of more than 70% in two or more major epicardial coronary arteries. Exclusion criteria included left main coronary stenosis ≥ 50%, class III or IV heart failure, presentation with myocardial infarction (MI), prior CABG or valve surgery, prior PCI within 6 months, prior stroke, dementia, and life expectancy < 5 years. Patients were randomized 1:1 to either PCI or CABG. Minimum follow-up for all patients was 2 years. Sirolimus-eluting (51%) and paclitaxel-eluting (43%) stents were the predominant types of drug-eluting stents (DES) used in the trial. The study protocol recommended that only one type of DES should be used in a given patient. The use of abciximab was recommended for patients undergoing PCI. The use of dual antiplatelet therapy with aspirin and clopidogrel was recommended for at least 12 months after PCI. For CABG surgery, arterial revascularization was encouraged. Medical therapy for all patients was recommended, according to guidelines. The primary outcome was a composite of death from any cause, nonfatal MI, and stroke.

Patients were randomized to receive CABG (n = 947) or PCI with DES (n = 953). Baseline demographics were similar between groups with a mean age of 63 years, mean LVEF of 66%, mean Euroscore of 2.7, mean syntax score 26, and triple vessel disease in 83%. The primary endpoint occurred in 26.6% of PCI patients and 18.7% of CABG patients after 5 years (P = 0.005). The rates of death (16.3% vs 10.9%; P = 0.049) and MI (13.9% vs 6.0%; P < 0.001) were higher in the PCI group compared to CABG, whereas stroke was higher in the CABG group compared to PCI (5.2 vs 2.4%; P = 0.03).

Several prespecified secondary endpoints were examined. The rate of cardiovascular death (63.7% of all deaths) did not differ significantly between the two study groups (P = 0.12). One year after the procedure, there was a significant difference in rates of major adverse cardiovascular and cerebrovascular events (MACCE: primary endpoint + repeat revascularization), with 16.8% in the PCI group vs 11.8% in the CABG group (P = 0.004). This difference was attributable to increased revascularization events in the PCI group (12.6% vs 4.8%; P < 0.001). In pre-specified subgroup analyses, there was consistency in benefit from CABG across all subgroups. In safety analyses, there were no differences between groups in 30-day rates of bleeding, but acute renal failure requiring hemodialysis within 30-day rates of the index procedure occurred in one patient in the PCI group vs eight patients in the CABG group (P = 0.02). The authors conclude that for patients with diabetes and advanced CAD, CABG was superior to PCI in that it significantly reduced rates of death and MI with a higher rate of stroke.


The FREEDOM trial is a large, well-designed study that shows superiority of CABG over PCI for diabetics with multivessel CAD for the combined endpoint of death, MI, and stroke. This confirms what most of us have been telling our patients: that CABG has better long-term outcomes in diabetics, albeit at some increased early risk. The results of this study are consistent with most other studies comparing CABG and PCI in diabetics. There is consistently a lower rate of MI with CABG and a lower rate of stroke with PCI. The mortality data are conflicting. Another smaller study, the CARDia trial, which also randomized diabetics to PCI or CABG, confirmed the higher MI and lower stroke rate with PCI, but showed no significant difference in mortality between groups. Both the authors and the accompanying editorial to the FREEDOM trial discounted this trial because it was smaller (510 patients). However, had the same numeric outcomes occurred in this trial, despite its larger size, the mortality between groups would not have been significantly different. Furthermore, the rate of cardiovascular death in the FREEDOM trial was not different between groups, but there is no obvious explanation for this. For these reasons, I interpret the mortality benefit of CABG over PCI at 5 years with caution. However, the Kaplan-Meier mortality curves start to separate in favor of CABG after about 4 years, and I think we will see the mortality curves continue to separate with time, favoring CABG. The data from this and other trials are overwhelmingly clear that CABG reduces the rate of MI, and PCI has lower rates of stroke.

There are some important factors to consider in interpreting this dataset. First, this was performed with contemporary medical therapy and surgical practice, but with first-generation drug-eluting stents (DES). Neither of the DES used in 94% of patients is on the market anymore, and these stents have been supplanted by better, second-generation DES. Recent data suggest significantly lower rates of MACE with second-generation DES over first-generation DES. Second, there were significant exclusion criteria that omitted the higher-risk patients: left main coronary stenosis ≥ 50%, class III or IV heart failure, presentation with MI, prior CABG or valve surgery, prior PCI within 6 months, and prior stroke. The results may not be generalizable to the high-risk groups omitted from the trial. Finally, we await more data from this trial, which will be forthcoming in subsequent manuscripts, such as the cost-effectiveness analysis. Despite these factors, the FREEDOM trial confirms that CABG remains the treatment of choice for patients with diabetes and multivessel CAD. Longer-term follow-up may show even greater benefit in favor of CABG.