ART

Published

Taggart DP, et al. "Bilateral versus Single Internal-Thoracic-Artery Grafts at 10 Years". The New England Journal of Medicine. 2019. 380(5):437-446.
PubMed • Full text • PDF

Clinical Question

In patients with multivessel coronary disease undergoing coronary artery bypass grafting (CABG), is bilateral internal mammary artery (BIMA) grafting superior to single IMA grafting with regard to overall mortality?

Bottom Line

In patients with multivessel coronary disease selected for surgical revascularization, bilateral internal mammary artery (BIMA) grafting resulted in similar 10 year mortality as compared to single IMA grafting. There was a 1.6% absolute increase in sternal wound complications using BIMA.

Major Points

On the basis of multiple randomized trials including FREEDOM and BARI-2D, coronary artery bypass grafting (CABG) has become the standard of care for patients with multivessel coronary artery disease (CAD), particularly in the presence of diabetes. Use of the left internal mammary artery (IMA) as a graft conduit is almost always part of the surgical revascularization plan, particularly given evidence suggesting improved long-term patency as compared to vein grafts.

Although less commonly utilized, the right IMA is also a potential graft conduit. Observational studies have shown lower long-term mortality when both the right and left IMAs are used, which may be related to improved long-term patency of the right IMA as compared to vein grafts. However, use of the bilateral IMAs is more technically challenging, and may be associated with a greater rate of sternal wound dehiscence. Thus, a randomized comparison of bilateral IMA (BIMA) with traditional single IMA for coronary artery revascularization was needed.

The 2019 Arterial Revascularization Trial (ART) randomized 1548 patients with multivessel CAD to surgical bypass with single IMA versus BIMA and assessed for a primary outcome of death. At 10 years, there was no significant difference in mortality between groups randomized to single IMA versus BIMA. There was an absolute 1.6% increase in sternal wound complications in the BIMA group. Importantly, use of other arterial grafts (e.g., radial) was not constrained, and 21.8% of patients randomized to single IMA grafting also had a radial artery graft placed. In an as-treated analysis considering only multiple arterial grafts (including radial) versus single arterial grafts, there was an absolute 4.5% reduction in mortality associated with use of multiple arterial grafts, although this analysis should be considered hypothesis-generating only.

The results of ART suggest that there is no clear advantage to the use of BIMA as compared to traditional single IMA bypass as a routine revascularization strategy, and there is a modest increase in sternal wound complications. Since use of non-IMA arterial grafts was not constrained, and over 20% of patients randomized to single IMA grafting also had a radial artery graft placed, it remains possible that maximal use of arterial grafts, rather than IMA grafts per se, may be associated with improved outcomes, as suggested by an as-treated analysis in ART. Further research is needed to address this consideration.

Guidelines

As of January 2019, no guidelines have been published that reflect the results of this trial.

Design

• Multicenter, randomized, open-label trial
• N=3102
  • Single IMA (N=1554)
  • Bilateral IMA (N=1548)
• Setting: 28 hospitals in seven countries
• Enrollment: June 2004 - December 2007
• Duration of follow-up: 10 years
• Analysis: Intention-to-treat
• Primary Outcome: Death

Population

Inclusion Criteria

• CABG patients with multivessel coronary artery disease (including urgent and off-pump CABG patients)

Exclusion Criteria

• Single graft
• Redo CABG
• Evolving myocardial infarction
• Concomitant valve surgery

Baseline Characteristics

From the single IMA graft group.

• Demographics: age 63.5 years, male 86.1%, white 92.1%
• Comorbidities: BMI 28.1, smoker 13.8%, DM 76.6%, HTN 78.3%, HLD 93.2%, PAD 7.6%, TIA 3.7%, stroke 3.1%, previous MI 43.9%, previous PCI 16.0%
• Heart failure: NYHA I 31.0%, NYHA II 48.1%, NYHA III 16.9%, NYHA IV 3.9%
• Angina: none 8.2%, CCS I 22.8%, CCS II 38.5%, CCS III 22.6%, CCS IV 7.9%

Interventions

• Patients randomized 1:1 to single or bilateral IMA grafting
• Randomization stratified by center
• Surgery recommended to be performed within 6 weeks of randomization
• Bilateral IMA group
  • Received bilateral IMA grafts to the two most important coronary arteries on the left side, with supplemental vein grafts or radial artery grafts as clinically indicated
  • IMA grafts could be used as composite grafts to each other, as long as one remained in situ
  • Anastomosis of an IMA graft to the right coronary artery was not permitted due to concerns regarding long-term patency
• Single IMA group
  • Received single IMA graft to the left anterior descending artery plus supplemental vein or radial artery grafts as clinically indicated
• Surgeons could participate in the trial only if their experience included 50 or more operations using bilateral IMA grafts
• Standard methods for anesthesia and myocardial protection were used according to local practice
• Evidence-based medical therapies were encouraged in all patients
• Data gathered at participating sites by means of annual telephone calls or hospital visits
• Serious adverse events were reported by investigators on specific forms
• Vital status was obtained by contact with family doctors and central registers where available
• All outcome events underwent adjudication

Outcomes

Comparisons are BIMA versus single IMA grafting

Primary Outcomes

All-cause mortality

315 (20.3%) vs. 329 (21.2%); HR 0.96 (95% CI 0.82-1.12); p = 0.62

Secondary Outcomes

Death, myocardial infarction, stroke

385 (24.9%) vs. 425 (27.3%); HR 0.90 (95% CI 0.79-1.03)

Myocardial infarction

71 (4.6%) vs. 78 (5.0%); HR 0.92 (95% CI 0.66-1.26)

Stroke

57 (3.7%) vs. 76 (4.9%); HR 0.75 (95% CI 0.53-1.06)

Safety Outcomes

Major bleeding

52 (3.4%) vs. 48 (3.1%); HR 1.09 (95% CI 0.74-1.61)

Repeat revascularization

159 (10.3%) vs. 156 (10.0%); HR 1.02 (95% CI 0.83-1.26)

Sternal wound complication

54 (3.5%) vs. 30 (1.9%); HR 1.81 (95% CI 1.16-2.81)

Sternal wound reconstruction

31 (2.0%) vs. 10 (0.6%); HR 3.11 (95% CI 1.53-6.32)

Subgroup Analyses

• There were no significant interactions between trial group and events according to age, diabetes, off-pump versus on-pump surgery, radial artery grafting, number of grafts, and ejection fraction.

Criticisms

• Only 84% of patients randomized to BIMA actually received bilateral IMA grafts. Although this may bias results against a benefit to BIMA, there was still no suggestion of benefit in a per-protocol analysis (in which only patients who received BIMA were considered).
• Various aspects of surgical technique (graft strategy, method of harvest, on-pump versus off-pump) were not controlled. Thus, it is possible that the results are biased by random imbalances with unknown effects on clinical outcome.
• Open-label trial design allows for the potential for bias in assessment of outcomes

Funding

• Study funded by the British Heart Foundation, the UK Medical Research Council, and the National Institute of Health Research Efficacy and Mechanistic Evaluation Program
• The institution and funders had no role in the design or conduct of the trial, in the analysis of the data, or in the writing of the manuscript or the decision to submit the manuscript for publication