MUNICH — Will patients who need coronary revascularization rather face a small increased risk of a stroke or a larger risk for a repeat procedure within a few months?
That is the decision facing patients with complex coronary disease, based on results from the largest and most tightly controlled study to ever compare percutaneous coronary stenting and coronary surgery.
“The risk of death, stroke, and MI is identical” between coronary stenting and surgery during the first year following intervention, “but the risk for more reintervention with PCI [percutaneous coronary intervention] is real,” Dr. Patrick W. Serruys said at the annual congress of the European Society of Cardiology.
Many other patients who need revascularization won't have a choice, based on findings from the Synergy Between Percutaneous Coronary Intervention With Taxus and Cardiac Surgery (SYNTAX) study. Out of 3,075 enrolled patients with either left main disease or triple vessel disease, 1,275 (41%) were judged by a team of cardiologists and cardiac surgeons to have no alternative in their revascularization treatment because of the complexity of their disease (including chronic total occlusion), comorbidities, or other factors that ruled out either surgery or stenting. For 1,077 of the nonrandomized patients (84%), bypass surgery was the only recourse; for the other 198 nonrandomized patients (16%), surgery was not feasible and so they had to be treated by PCI.
The other 1,800 patients (59%) in the study were deemed equally amenable to stenting or surgery and were randomized.
Although the results from both the randomized and registry arms highlighted recent progress toward better outcomes by both interventionalists and surgeons, the findings “probably will not change the number” of patients in routine practice who undergo stenting or have surgery, commented Dr. Spencer B. King III, an interventional cardiologist and executive director of academic affairs at Saint Joseph's Health System in Atlanta.
“About 80% of the types of patients in SYNTAX now go to surgery in the United States, and my guess is that this will stay the same,” Dr. King said in an interview. “The majority of these patients are seen by interventional cardiologists, and they are the biggest referrers of patients to surgeons. Surgeons do patients like these”—patients with left main or triple vessel disease—“all the time. It's bread-and-butter surgery,” Dr. King said.
But “these are hard cases for interventionalists. They take hours, and most interventionalist cardiologists don't want to do them,” commented Dr. W. Douglas Weaver, chief of cardiology at Henry Ford Hospital in Detroit.
A limitation of the study is that patients were followed for just 1 year. The new data “add to the discussion of using PCI for left main disease, but 1 year of follow-up is not very long to say that survival in patients with left main disease” is as good as in patients treated with surgery, Dr. King said. “The danger is that patients who develop a severe restenosis in their left main may die.”
SYNTAX was done at 62 European and 23 U.S. centers. Patients who entered the randomized part of the study had an average age of 65, and about 28% had diabetes. About two-thirds of patients had triple vessel disease, and about a third had a significant left main stenosis (patients with left main disease could have additional stenoses in one, two, or three other coronary arteries). All lesions were previously untreated, none of the patients had an acute MI, and none of the bypass surgery patients received concomitant cardiac surgery. The patients treated with stents received an average of 4.6 stents each. Only paclitaxel-eluting coronary stents were used. Although the study exclusively used Taxus stents, Dr. Serruys and the study cochair, Dr. Friedrich W. Mohr, reported no conflicts of interest.
After 1 year, the combined rate of death, nonfatal cerebrovascular accident (stroke), or nonfatal MI was virtually identical: 7.6% in 903 PCI patients, compared with 7.7% in the 897 patients treated with coronary artery bypass grafting (CABG). The only statistically significant difference between the two groups was a 2.2% rate of stroke in the CABG patients, compared with a 0.6% rate in the PCI patients. (See box.)
The study's primary end point combined the rate of these three “irreversible” events with the fourth major outcome, need for revascularization. The total for all four types of outcomes after 1 year was 12.1% in the CABG patients and 17.8% in the PCI patients, a statistically significant difference. This rate was also used to judge whether PCI was noninferior to CABG. The prespecified, noninferiority limit was a difference of less than 6.6% between the two treatments. Because the 95% confidence range for the quadruple end point was an excess as high as 8.3% in patients having PCI, the test for noninferiority was not met and so technically the results did not prove that PCI is not inferior to CABG. But Dr. Serruys acknowledged that having a combined end point that included revascularization was a controversial decision.