The New England Journal of Medicine March 05, 2009

Percutaneous Coronary Intervention versus Coronary-Artery Bypass Grafting for Severe Coronary Artery Disease (SYNTAX)

Patrick W. Serruys et al.

Bottom Line

In patients with severe three-vessel or left main coronary artery disease, CABG showed superior clinical outcomes compared to PCI with first-generation drug-eluting stents at 12 months, primarily driven by a significantly lower need for repeat revascularization, though PCI resulted in fewer strokes.

Key Findings

1. PCI failed to meet the criteria for non-inferiority compared to CABG for the primary composite endpoint of MACCE at 12 months, which occurred in 17.8% of the PCI group versus 12.4% of the CABG group (P=0.002).

2. The difference in MACCE was primarily driven by a significantly higher rate of repeat revascularization in the PCI arm (13.5% vs. 5.9% for CABG, P<0.001).

3. Stroke was significantly more frequent in the CABG group compared to the PCI group at 12 months (2.2% vs. 0.6%, P=0.003).

4. The rates of death from any cause or myocardial infarction were statistically similar between the PCI and CABG groups (death: 4.4% vs. 3.5%, P=0.37; MI: 4.8% vs. 3.3%, P=0.11).

5. Subgroup analysis utilizing the SYNTAX score demonstrated that in patients with low scores (0-22), MACCE rates were similar between groups, whereas in patients with intermediate (23-32) or high (≥33) scores, CABG was associated with a significantly lower rate of MACCE.

Study Design

Design

RCT

Open-Label

Sample

1,800

Patients

Duration

12 mo

Median

Setting

Multicenter, multinational

Population Patients with previously untreated three-vessel or left main coronary artery disease (or both) who were deemed by a Heart Team to be suitable candidates for either CABG or PCI

Intervention Percutaneous coronary intervention (PCI) using first-generation paclitaxel-eluting stents (TAXUS Express)

Comparator Coronary artery bypass grafting (CABG)

Outcome Major adverse cardiac or cerebrovascular events (MACCE), a composite of death from any cause, stroke, myocardial infarction, or repeat revascularization at 12 months

Study Limitations

• The trial utilized first-generation paclitaxel-eluting stents, which have since been superseded in contemporary practice by safer, thinner-strut, and more efficacious second-generation drug-eluting stents.

• The open-label design of the trial introduces inherent ascertainment and performance biases, particularly for softer endpoints such as the decision to pursue repeat revascularization.

• The 12-month follow-up for the primary report is relatively short for comparing long-term revascularization strategies, as mortality differences often take years to manifest.

• The prespecified non-inferiority margin was relatively wide, yet PCI still failed to meet it, highlighting the significant performance gap at the time.

• Advances in surgical technique, such as the routine use of multiple arterial conduits, and improvements in adjunctive medical therapy since 2009 may limit the direct applicability of these specific outcome rates to modern practice.

Clinical Significance

The SYNTAX trial fundamentally shifted the revascularization paradigm by solidifying CABG as the standard of care for patients with highly complex coronary artery disease (intermediate to high SYNTAX scores) due to superior freedom from repeat revascularization. Conversely, it validated PCI as a safe and acceptable alternative for patients with less complex anatomy (low SYNTAX scores) or isolated left main disease. Crucially, it established the 'Heart Team' approach—requiring joint consensus between an interventional cardiologist and a cardiac surgeon—and cemented the SYNTAX score as an indispensable anatomical risk-stratification tool now embedded in major international societal guidelines.

Historical Context

Historically, CABG was the unquestioned gold standard for severe multivessel and left main coronary artery disease. However, the advent of bare-metal stents and subsequently first-generation drug-eluting stents dramatically reduced the invasiveness and restenosis rates of PCI, expanding percutaneous therapy into increasingly complex anatomies. The SYNTAX trial was a landmark milestone as the first large-scale, prospective, randomized trial to specifically pit CABG against DES-based PCI in the most complex coronary patients using a broad 'all-comers' screening approach. It introduced the SYNTAX score—a quantitative angiographic tool that revolutionized clinical decision-making by matching anatomical complexity with the optimal revascularization strategy.

Guided Discussion

High-yield insights from every perspective

Med Student

Medical Student

What is the underlying anatomical and pathophysiological rationale for why CABG traditionally offers more durable outcomes with less need for repeat revascularization compared to PCI in severe three-vessel coronary artery disease, as seen in the SYNTAX trial?

Key Response

CABG bypasses the proximal diseased segments entirely by creating a new conduit (like the LIMA to the LAD) to the healthy distal vessel, rendering future proximal progression of native atherosclerosis clinically irrelevant. In contrast, PCI treats only the specific focal stenotic lesion, leaving the rest of the native vessel vulnerable to neoatherosclerosis, stent restenosis, and the progression of disease in adjacent segments.

Resident

When counseling a 65-year-old patient with complex three-vessel disease and a high SYNTAX score (>33), how do you clinically weigh and communicate the competing perioperative risks and long-term benefits of CABG versus PCI?

Key Response

Residents must practice shared decision-making. High SYNTAX scores strongly favor CABG due to significant long-term mortality and revascularization benefits, but the upfront risk of stroke is higher with surgery (often related to aortic manipulation and clamping). Discussing these specific trade-offs—higher upfront procedural stroke/recovery risk with CABG versus the high likelihood of requiring repeat procedures with PCI—is essential for informed consent.

Fellow

The SYNTAX trial analyzed patients by SYNTAX score tertiles and disease type (Left Main vs. 3-Vessel). How does the presence of isolated Left Main disease with a low-to-intermediate SYNTAX score alter the risk-benefit ratio of PCI versus CABG compared to diffuse 3-vessel disease?

Key Response

Fellows need to recognize critical subgroup nuances. In SYNTAX, PCI outcomes in left main disease with low/intermediate SYNTAX scores (<=32) were comparable to CABG in terms of MACCE, largely because left main lesions are often focal and can be effectively stented. Conversely, severe diffuse 3-vessel disease almost universally favored CABG. This specific finding drove subsequent major trials (like EXCEL and NOBLE) specifically evaluating PCI in left main disease.

Attending

The SYNTAX trial was instrumental in establishing the multidisciplinary 'Heart Team' approach. From an attending perspective, how do we prevent the Heart Team discussion from becoming a mere formality and ensure unbiased decision-making when modern procedural technology (e.g., 3rd-generation DES, intracoronary imaging) has rapidly outpaced the original 1st-generation stent data from the trial?

Key Response

Attendings must contextualize older trial data with modern tools while maintaining multidisciplinary integrity. The Heart Team is designed to curb selection bias and provide optimal care, but it requires active, equitable collaboration between interventionalists and surgeons. Balancing the historical standard of CABG durability with the genuine advances in modern PCI safety requires nuanced, patient-centered judgment rather than rigid adherence to outdated stent outcome data.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD

The SYNTAX trial utilized a non-inferiority design for its primary composite endpoint of MACCE, which PCI failed to meet. How does the inclusion of 'repeat revascularization'—a highly subjective, physician-driven event—in a composite primary outcome complicate the interpretation of an unblinded trial comparing a percutaneous versus a surgical intervention?

Key Response

Methodologically, composite endpoints in unblinded trials are vulnerable to bias if the individual components have different thresholds for occurrence. PCI leaves native vessels amenable to relatively easy repeat angiography and stenting, while repeat CABG is a formidable, high-risk surgery. This asymmetry in the clinical threshold to re-intervene can disproportionately inflate the MACCE rate in the PCI arm, threatening the internal validity of the effect size.

Journal Editor

If reviewing this manuscript today, how would you critique the protocol's exclusive use of the TAXUS paclitaxel-eluting stent, given that it was already becoming obsolete in favor of second-generation DES during the study's enrollment, and how does this affect the manuscript's generalizability?

Key Response

Editors must assess whether a trial's findings have a viable clinical 'shelf-life.' The use of first-generation DES, which had notoriously higher rates of stent thrombosis, target lesion revascularization, and delayed healing compared to newer everolimus/zotarolimus stents, inherently handicapped the PCI arm. A critical reviewer would flag this as a major threat to the external validity and long-term applicability of the study's conclusions regarding PCI efficacy.

Guideline Committee

Based on the SYNTAX trial and its long-term follow-ups, current ACC/AHA guidelines heavily rely on angiographic SYNTAX score cutoffs to issue Class I vs III recommendations for CABG vs PCI. Should these strict anatomical cutoffs be revised to incorporate 'Functional SYNTAX' scores utilizing modern physiology (FFR/iFR) and intravascular imaging?

Key Response

Current guidelines strongly recommend CABG (Class I) and advise against PCI (Class III) for high SYNTAX scores (>33). However, the committee must evaluate whether anatomical grading alone remains sufficient. Integrating physiological lesion assessment (FFR/iFR) often downgrades the functional severity of multivessel disease, potentially shifting patients into lower risk categories where modern, imaging-optimized PCI achieves outcomes comparable to CABG, thereby necessitating a modern update to the guidelines.

Clinical Landscape

Noteworthy Related Trials

2012

FREEDOM Trial

n = 1,900 · NEJM

Tested

PCI with drug-eluting stents

Population

Patients with diabetes and multivessel coronary artery disease

Comparator

CABG

Endpoint

Composite of death from any cause, nonfatal myocardial infarction, or nonfatal stroke

Key result: CABG was superior to PCI in reducing the rates of death and myocardial infarction, though with a higher rate of stroke.

2016

EXCEL Trial

n = 1,905 · NEJM

Tested

PCI with everolimus-eluting stents

Population

Patients with left main coronary artery disease and low or intermediate SYNTAX scores

Comparator

CABG

Endpoint

Composite of death from any cause, stroke, or myocardial infarction at 3 years

Key result: PCI was noninferior to CABG with respect to the primary composite endpoint at 3 years.

2016

NOBLE Trial

n = 1,201 · Lancet

Tested

PCI with biolimus-eluting stents

Population

Patients with unprotected left main coronary artery disease

Comparator

CABG

Endpoint

Composite of all-cause mortality, non-procedural myocardial infarction, any repeat coronary revascularization, or stroke at 5 years

Key result: CABG was significantly better than PCI for the primary composite endpoint, driven mostly by higher rates of non-procedural MI and repeat revascularization in the PCI group.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis