Review of the SYNTAX trial
PCI vs bypass surgery for coronary artery disease
Background: After its introduction in 1968, the use of coronary artery bypass grafting (CABG) in patients with coronary artery disease grew rapidly. Although early trial results were mixed (see the Veterans Administration Cooperative study, the European Coronary Surgery Study and the CASS study), advancements in surgical techniques and the use of arterial conduits improved survival and reduced graft occlusion rates. In 1977, percutaneous coronary intervention (PCI) was introduced, and subsequent improvements in PCI techniques and stents have led to fewer complications and better stent patency. Large trials comparing CABG to PCI using drug-eluting stents were lacking.
The Synergy between PCI with Taxus and Cardiac Surgery (SYNTAX) trial sought to assess the optimal revascularization strategy for patients with left main or three-vessel coronary artery disease.
Patients: Eligible patients had three-vessel or left main coronary artery disease. Patients had to have 50% or more stenosis in the target vessels and either chest pain or evidence of myocardial ischemia. Exclusion criteria were previous PCI or CABG, acute myocardial infarction or the need for concomitant cardiac surgery.
Baseline characteristics: The trial randomized 1,800 patients – 903 randomized to PCI and 897 to CABG.
The average age of patients was 65 years and 78% were men. The average body mass index was 28 kg/m2. Approximately 67% had hypertension, 78% had hyperlipidemia, 25% had diabetes, 33% had prior myocardial infarction, 5% had congestive heart failure, and 20% were current smokers.
The number of target lesions was approximately 4 in both groups. The SYNTAX score was 28.4 in the PCI group and 29.1 in the CABG group.
Note to readers: The SYNTAX score is an anatomical score to grade the complexity of coronary artery disease with higher scores indicating more complex disease. A score of 22 or less is considered low whereas a score of 33 or more is considered high (this scoring was a prespecified secondary endpoint).
Procedures: Patients were assigned in a 1:1 ratio to undergo CABG or PCI using Taxus Express paclitaxel-eluting stents (drug-eluting stent). Randomization was stratified by the presence or absence of left main disease and diabetes.
The intent was to achieve complete revascularization in all coronary arteries of at least 1.5 mm diameter with 50% or more stenosis. The techniques for CABG and PCI and post procedural medications were chosen based on local practices.
Endpoints: The primary endpoint was a composite of death from any cause, stroke, myocardial infarction, or repeat revascularization up to 12 months after randomization.
Analysis was performed based on the intention-to-treat principle. The primary analysis was a non-inferiority analysis. The estimated sample size was 1800 patients assuming 13.2% event rate in the CABG group, 14.0% in the PCI group and 6.6% absolute risk difference non-inferiority margin. This sample size would provide 96% power at 5% one-sided alpha.
Results: Revascularization was complete in 56.7% of the patients in the PCI group and 63.2% in the CABG group. In the CABG group, one or more arterial grafts were used in 97.3% of the patients. In the PCI group, the average number of stents implanted per patient was more than four.
At 12 months, the primary outcome was lower in the CABG group (12.4% vs 17.8%, RR with PCI: 1.44, 95% CI: 1.15 – 1.81; p= 0.002). The absolute difference was 5.4% with an upper bounds of the 95% CI of 8.3%; thus not meeting non-inferiority. Stroke was lower with PCI (0.6% vs 2.2%; p= 0.003) while repeat revascularization was lower with CABG (5.9% vs 13.5%; p< 0.001). There was no significant difference in death from any cause (4.4% with PCI vs 3.5% with CABG; p= 0.37) or myocardial infarction (4.8% with PCI vs 3.3% with CABG; p= 0.11).
There was a significant interaction based on the SYNTAX score where patients with high SYNTAX scores (>/=33) had significant benefit with CABG while patients with low or intermediate SYNTAX scores did not (14.7% with CABG vs 13.6% with PCI for low SYNTAX score, 12.0% with CABG vs 16.7% with PCI for intermediate SYNTAX score, and 10.9% with CABG vs 23.4% with PCI for high SYNTAX score; p for interaction= 0.01).
In the subgroup of patients with left main disease, the rate of the primary outcome at 12 months was not significantly different between both treatment groups (13.7% with CABG vs 15.8% with PCI; p= 0.44). In patients with three-vessel but no left main disease, the primary outcome was significantly higher in the PCI group (19.2% vs 11.5%; p< 0.001).
Stent thrombosis was 3.2% in the PCI group and graft occlusion was 3.1% in the CABG group.
Conclusion: In patients with stable three-vessel or left main coronary artery disease, PCI did not meet non-inferiority compared to CABG in reducing the composite endpoint of death from any cause, stroke, myocardial infarction, or repeat revascularization over 12 months of follow up. The number of patients needed to treat with CABG to prevent one primary endpoint was approximately 19 patients, which was mainly accounted for by repeat revascularization. There was no significant difference in death or myocardial infarction and stroke was higher in the CABG group.
The subgroup analysis based on SYNTAX score demonstrated a strong signal that CABG is more effective than PCI in patients with complex coronary artery disease.
In patients with stable coronary artery disease, meeting the inclusion criteria of this trial, the risk of repeat revascularization with PCI should be carefully weighed against the increased stroke risk and invasiveness of CABG. Considerations regarding patient selection are crucial. It is reasonable to anticipate that less robust patients will have higher morbidity with CABG versus PCI.