The trusted source for
healthcare information and
By Jeffrey Zimmet, MD, PhD
Associate Professor of Medicine, University of California, San Francisco; Director, Cardiac Catheterization Laboratory, San Francisco VA Medical Center
Dr. Zimmet reports no financial relationships relevant to this field of study.
SYNOPSIS: Five-year results from the EXCEL trial, which randomized 1,905 patients with left main disease to coronary artery bypass grafting or percutaneous coronary intervention, revealed no significant difference in the primary composite outcome of death, stroke, or myocardial infarction.
SOURCE: Stone GW, Kappetein AP, Sabik JF, et al. Five-year outcomes after PCI or CABG for left main coronary disease. N Engl J Med 2019;381:1820-1830.
In late 2016, the EXCEL trial investigators, who randomized patients with left main coronary artery disease to coronary artery bypass grafting (CABG) or percutaneous coronary intervention (PCI), published the primary analysis of this important study. The authors reported three-year outcomes, which showed noninferiority of PCI to CABG with respect to a composite of death, myocardial infarction (MI), and stroke at three years.1 Why do the five-year data merit publication?
EXCEL was an international, multicenter, open-label trial that included patients with significant left main disease and a consensus among the heart team that either PCI or CABG was a valid consideration for revascularization. The design of the trial was such that subjects were intended to present with coronary disease that was of low or intermediate complexity, as defined by a SYNTAX score of ≤ 32. Ultimately, 1,905 patients were randomly assigned at 126 sites to PCI (948 patients) or to CABG (957 patients). Just over half presented with stable angina, with the remainder presenting with acute coronary syndromes. SYNTAX scores in both groups averaged just over 20, with most patients categorized as low complexity. Eighty percent of left main lesions involved the distal bifurcation. Most patients underwent the assigned treatments (942 of 948 PCI patients; 940 of 957 CABG patients). The primary outcome of death, stroke, or MI at five years occurred in 22% of patients in the PCI group and in 19.2% of patients in the CABG group (difference, 2.8 percentage points; P = 0.13). In contrast, a secondary composite outcome that included ischemia-driven revascularization was more common in the PCI group (31.3% of the PCI group vs. 24.9% of the CABG group; odds ratio, 1.39; P = 0.002).
Among the individual secondary outcomes, ischemia-driven revascularization was more common in the PCI group at five years, whereas cerebrovascular events (although driven primarily by transient ischemic attack rather than by stroke) were seen more commonly among CABG patients. All-cause death at five years accounted for 13% of patients in the PCI group and 9.9% of patients in the CABG group. However, this difference was driven by noncardiovascular deaths (primarily infection and cancer), and its significance is unclear.
The authors concluded that among patients with left main disease of low or intermediate complexity randomized to PCI or CABG, there was no significant difference between the groups at five years in terms of the composite of death, stroke, or MI.
The interesting part of this paper is not to be found in the abstract. Indeed, the simple tally of the primary composite outcome was not significantly different between the PCI and CABG groups at five years, similar to the result of the initial publication at three years. However, when the results are analyzed during distinct periods following revascularization, a different picture emerges.
Between zero and 30 days after revascularization, PCI showed a distinct advantage over CABG regarding the composite of death, stroke, and MI, with a hazard ratio (HR) of 0.61 (95% confidence interval [CI], 0.42-0.88). This difference was driven primarily by MI, with no significant difference in the other components of the primary endpoint (including stroke, surprisingly). Between 30 days and one year, the two groups are comparable (HR, 1.07; 95% CI, 0.68-1.70). From one year to five years, the data suggest an advantage to CABG (HR comparing PCI to CABG, 1.61; 95% CI, 1.23-2.12). At this later point, the main driver appears to be spontaneous MI, which itself carries a HR of 2.16 (95% CI, 1.27-3.67).
This description of distinct periods of relative risk, wherein the early benefits of PCI from reduced procedural risk are offset by more events during later follow-up, matches a common narrative describing bypass surgery as a preventive measure against future MI.
However, the degree to which this relates to the procedure itself is in question. For example, it is notable that in this randomized trial, medical therapy differed significantly between the CABG and PCI groups. Even at five years, P2Y12 inhibitor use was three-fold higher in the PCI group, while beta-blockers, anti-arrhythmic agents, and oral anticoagulants were significantly more common among CABG patients.
In the end, longer-term data will be required to answer this question more fully. The authors suggested 10-year follow-up or longer may be required to fully characterize the long-term benefits and vulnerabilities of these two procedures.
Financial Disclosure: Clinical Cardiology Alert’s Physician Editor Michael H. Crawford, MD, Peer Reviewer Susan Zhao, MD, Nurse Planner Aurelia Macabasco-O’Connell, PhD, ACNP-BC, RN, PHN, FAHA, Editor Jonathan Springston, Editor Jason Schneider, Editorial Group Manager Leslie Coplin, and Accreditations Manager Amy M. Johnson, MSN, RN, CPN, report no financial relationships relevant to this field of study.