Coronary Disease2019New England Journal of Medicine

EXCEL

Evaluation of XIENCE Versus Coronary Artery Bypass Surgery for Effectiveness of Left Main Revascularization

Sample Size

1,905

Study Design

Multicenter, international, randomized, open-label, noninferiority trial

Year Published

2019

Clinical Question

Is PCI with everolimus-eluting stents noninferior to CABG for patients with left main coronary artery disease of low or intermediate anatomic complexity?

Population

Patients with unprotected left main coronary artery disease and a SYNTAX score of 32 or less at 126 sites in 17 countries.

Intervention

PCI with everolimus-eluting stents (XIENCE)

Control

Coronary artery bypass grafting (CABG)

Primary Endpoint

Composite of death from any cause, stroke, or myocardial infarction at 3 years (5-year results published in 2019)

Key Findings

At 3 years, the primary endpoint occurred in 15.4% of PCI vs 14.7% of CABG patients (p=0.02 for noninferiority).

At 5 years, the primary endpoint was 22.0% with PCI vs 19.2% with CABG (p=0.13), with no significant difference.

All-cause mortality at 5 years was numerically higher with PCI (13.0% vs 9.9%; p=0.11).

Repeat revascularization was more common with PCI (16.9% vs 10.0%; p<0.001) at 5 years.

Periprocedural MI definitions became a subject of major controversy, with different definitions yielding different conclusions.

Impact on Clinical Practice

EXCEL generated significant controversy in cardiovascular medicine. The 3-year primary results suggested noninferiority of PCI for left main disease with low-to-intermediate complexity, while the 5-year follow-up revealed diverging mortality curves that raised concerns about the durability of PCI outcomes. The controversy centered on the definition of periprocedural myocardial infarction. Using the protocol-specified universal definition, CABG appeared to have higher MI rates, while the more clinically relevant SCAI definition yielded different results. This methodological dispute led to an independent review and widespread debate about trial integrity and endpoint selection in cardiovascular trials. Despite the controversy, EXCEL contributed important data on left main PCI and reinforced the need for careful long-term follow-up. Most guidelines continue to recommend CABG as the preferred strategy for left main disease, while acknowledging that PCI is a reasonable alternative in selected patients with low anatomic complexity.

Guideline Impact

EXCEL contributed to the ACC/AHA Class IIa recommendation for PCI as an alternative to CABG in left main disease with low anatomic complexity (SYNTAX score ≤22). However, the ESC/EACTS downgraded PCI for left main to Class IIa with higher evidence thresholds, citing long-term mortality concerns.

Limitations

The definition of periprocedural MI was a major source of controversy and may have influenced the primary endpoint results.

Open-label design introduced potential bias in postprocedural management and event ascertainment.

5-year all-cause mortality trended higher with PCI, and longer follow-up showed progressive separation of survival curves favoring CABG.

Reviewed by Serrie Lico, MD

Chief Medical Officer

How Does This Trial Apply to Your Case?

Clinical trials inform guidelines, but every patient is unique. Our Heart Team evaluates how landmark evidence applies to your specific anatomy, risk profile, and preferences.

Get a Second Opinion

Related Trials

SYNTAX

2009 · New England Journal of Medicine

FREEDOM

2012 · New England Journal of Medicine

ISCHEMIA

2020 · New England Journal of Medicine

left main disease coronary artery disease cabg vs pci

Stay informed.

Expert cardiac surgery insights from the WhiteGloveMD Heart Team, delivered to your inbox.

No spam. Unsubscribe anytime. HIPAA-compliant.