Double artery grafts show no survival edge at 15 years, trial follow-up finds

For years, surgeons have debated whether using two internal thoracic (mammary) arteries instead of one during coronary bypass surgery buys patients longer life. The theory is appealing: arterial grafts stay open longer than vein grafts, so two should beat one. A 15-year follow-up published on 2 September in JAMA Cardiology — the longest look yet at the randomised evidence — finds no survival difference at all [s1].

What the trial did

This is a secondary analysis of an unblinded randomised trial conducted at 28 cardiac-surgery centres in seven countries between June 2004 and December 2007 [s1]. Patients scheduled for coronary artery bypass grafting (CABG) on clinical grounds were eligible; those needing only a single graft, those having concurrent valve surgery, and those with previous bypass surgery were excluded [s1]. All 3,102 patients in the original trial contributed data, and 3,042 completed the full 15-year follow-up [s1]. Their mean age was 64, and 446 (15%) were women [s1]. The primary outcome for this analysis was all-cause mortality at 15 years; a composite of death, heart attack or stroke was secondary [s1].

What it found

A total of 1,548 patients were randomised to bilateral grafts and 1,554 to single grafts [s1]. At 15 years, vital status was known for 3,039 patients — 98% of the trial [s1]. There were 585 deaths (37.8%) in the bilateral-graft group and 584 (37.6%) in the single-graft group, a hazard ratio of 1.00 (95% CI, 0.89–1.12; P = .97) — as close to no difference as a trial can produce [s1]. The secondary composite outcome told the same story: 43.9% versus 45.8%, hazard ratio 0.94 (95% CI, 0.85–1.05) [s1].

The complication in the result

The finding comes with a caveat the authors emphasise, and it is not a small one. The two groups did not stay cleanly separated. In the bilateral-graft group, 215 patients (14%) ended up receiving only a single arterial graft, while in the single-graft group, 359 patients (23%) also received a radial artery graft — itself a second arterial conduit [s1]. In other words, a substantial share of "single-graft" patients got multiple arterial grafts anyway, and some "bilateral" patients did not.

That crossover blunts the contrast the trial set out to test. An intention-to-treat analysis — which this is — keeps patients in their assigned group regardless of what they actually received, preserving randomisation but diluting the comparison when crossover is heavy [s1]. The authors are explicit that, because of these post-randomisation departures, the broader hypothesis that more arterial grafts improve outcomes "still needs to be tested" in new randomised trials [s1].

How to read it

The clean version of the takeaway is that, at 15 years, patients randomised to two mammary-artery grafts lived no longer than those randomised to one [s1]. The honest version adds that the trial's ability to answer the underlying surgical question was compromised by how often surgeons deviated from the assignment — so this is strong evidence against a large survival benefit from the strategy as randomised, but not a final verdict on multiple arterial grafting itself.

Why it matters

Bilateral grafting is technically more demanding and carries its own risks, such as impaired wound healing in the breastbone. If it does not extend life, that trade-off matters to how surgeons choose. This long follow-up pushes the burden of proof back onto proponents: absent a survival signal here, any benefit will have to be demonstrated in a trial that keeps its groups intact [s1].

This article is informational and does not constitute medical advice.

Sources

• [s1] Long-Term Outcomes Following Bilateral vs Single Internal Thoracic Artery Grafts: A Secondary Analysis of a Randomized Clinical Trial. JAMA Cardiology, 2 September 2026. https://doi.org/10.1001/jamacardio.2026.3232