乌鸦传媒

This article is a collaboration between 乌鸦传媒 and:

Bypass surgery is a better treatment choice than stenting for diabetics with multivessel coronary artery disease.

In a study of 1,900 patients, the 5-year rate of death, myocardial infarction, or stroke was 26.6% among patients who had arteries opened with stents versus 18.7% in the coronary artery bypass graft (CABG) group, which represented 200 events in the stent arm versus 146 in the surgery group (P=0.005), wrote Valentin Fuster, MD, PhD, of Mount Sinai School of Medicine in New York City, and colleagues in a study published online Nov. 4 by the New England Journal of Medicine.

The superiority of surgery was driven by more than a doubling in the 5-year MI rate in the stent arm -- 13.9% versus 6% (P<0.001) -- and a lower mortality rate at 5 years, 16.3% versus 10.9% (P=0.049).

Moreover, surgery bested stents regardless of disease severity as measured by the SYNTAX score, beginning with a 6% benefit in composite endpoint favoring CABG for low-risk patients, 10% for moderate risk, and 8% for high-risk patients.

"This will absolutely change clinical practice because this is a big difference," Fuster told 乌鸦传媒. In addition to the NEJM publication, the FREEDOM results were also reported at a late-breaking clinical trials session at the American Heart Association meeting in Los Angeles.

In an editorial that accompanied the study, Mark Hlatky, MD, of Stanford University in Palo Alto, Calif., echoed Fuster's assessment.

"As a cardiologist who does not perform either procedure, I find that the FREEDOM trial provides compelling evidence of the comparative effectiveness of CABG versus [percutaneous coronary intervention] in patients with diabetes and multivessel coronary artery disease," he wrote.

And the finding is hardly surprising since 17 years ago the National Heart, Lung, and Blood Institute warned cardiologists that patients who underwent bypass surgery had better survival rates than patients treated with PCI. Since that time, a number of studies, most notably SYNTAX, reached the same conclusion.

Yet, in the real world of clinical practice, "more patients with diabetes have undergone PCI rather than CABG to treat multivessel coronary disease," Hlatky wrote.

Asked to speculate why the use of PCI has continued to grow as demand for bypass surgery has declined in the face of evidence supporting the benefit of surgery, heart surgeon Timothy Gardner, MD, told 乌鸦传媒, that patients, families, and cardiologists "don't like surgery because it is traumatic. It is traumatic at every level -- patient, family, and cardiologist."

As a result, he said, there is a "a grudging reluctance to accept evidence that favors CABG."

But Gardner, who is medical director of Christiana Care's Center for Heart and Cardiovascular Health in Newark, Del., and a former president of the AHA, said the evidence from FREEDOM should not be ignored.

"I think this is a class I, evidence level A, finding," he said.

Another reason for the apparent disconnect between evidence and practice may be stroke risk.

In CABG versus PCI trials, stroke rate is higher in the surgery arm and this again was the case in FREEDOM.

At the 5-year mark, the stroke rate was significantly higher in the surgery arm -- 5.2% versus 2.4% -- but the number of strokes was low in both arms -- 20 in the stent group and 37 in the surgery arm (P =0.03). However, this P-value was calculated as "P=0.16 by the Wald test of the Cox regression estimate for the study-group assignment in 1,712 patients after adjustment for the average glucose level after the procedure."

And although CABG was clearly superior on the primary endpoint at 5 years and at 2 years, at 30 days just 26 stent patients reached the composite endpoint versus 42 patients in the surgery arm.

The Future Revascularization Evaluation in Patients with Diabetes Mellitus: Optimal Management of Multivessel Disease (FREEDOM) study enrolled 1,900 patients over a 5-year period from 2005 through 2010. Most patients were in the early 60s (mean age 63) and most of the patients (83%) were men.

The mean SYNTAX score (a measure of disease complexity) was 26.2±8.6. Patients randomized to the PCI group who did receive interventional treatment were given either sirolimus-eluting stents (Cypher) or paclitaxel-eluting stents (Taxus).

The median follow-up for all patients was 3.8 years.

As might be expected, antiplatelet therapy was more common among patients in the stent group, but use of cardiovascular medications including statins, beta-blockers, angiotensin-converting enzyme inhibitors, and angiotensin receptor blockers, was the same in both groups.

Fuster and colleagues noted a number of limitations for the study including the unblinded design of the trial, which may have caused some surgical patients to "have been treated differently on the basis of their surgical procedure."

Another limitation, they wrote, was a lack of statistical power in some of the prespecified subgroups.

From the American Heart Association: