WHAT THE STUDY ACTUALLY SAYS

In older heart-attack patients, stress-testing first did not match routine angiography

EVAOLD tested whether stress imaging could safely spare patients aged 80 and over an invasive procedure. It was stopped early for futility, and the selective strategy did not prove non-inferior.

Death, non-fatal heart attack or non-fatal stroke at one yearStress imaging-guided selective: 24.1%; Routine invasive: 20.7%0%15%30%Stress imaging-guided selective24.1%Routine invasive20.7%
Death, non-fatal heart attack or non-fatal stroke at one year
GroupValue (%)
Stress imaging-guided selective24.1
Routine invasive20.7
Death, non-fatal heart attack or non-fatal stroke at one year EVAOLD primary endpoint; the trial was stopped early for futility, so these are interim figures. Source: European Society of Cardiology

In patients aged 80 and over having the most common kind of heart attack, using a stress test to decide who needs invasive investigation did not prove as safe as sending everyone for routine angiography, according to the EVAOLD trial presented at ESC Congress 2026 [s1]. The trial was stopped early for futility, and its message is a cautious one: a routine invasive approach should not be replaced with a more selective, stress-test-first strategy in these patients on the strength of this evidence [s1].

The clinical problem is real. A heart attack without ST-segment elevation, or NSTEMI, is usually managed with angiography — threading a catheter to the coronary arteries, injecting contrast dye and reopening blockages where needed [s1]. In the oldest patients the balance is genuinely uncertain, because the procedure carries a higher risk of complications and randomised evidence of benefit in this group has been thin [s1].

What the trial tested

EVAOLD asked whether stress imaging could act as a gatekeeper — a non-invasive way to find the older patients most likely to benefit from an invasive strategy, and to spare the rest [s1]. Stress imaging assesses how far blood flow to the heart falls when the heart is made to work harder with a drug, using echocardiography or single-photon emission computed tomography [s1].

The trial, run at 25 French centres, enrolled patients aged 80 or older hospitalised with NSTEMI and randomly assigned them to one of two strategies [s1]. One group received stress imaging-guided selective invasive management: a stress test, then angiography only for those with moderate-to-severe disease. The other received routine invasive management: angiography for everyone, without a stress test [s1]. It was an open-label, non-inferiority design, registered as NCT03289728 [s2].

What happened

At the first prespecified interim analysis, after 587 of a planned 1,756 patients had been randomised, the trial was stopped for futility, because the conditional power to show non-inferiority had become low [s1]. In plain terms, the selective strategy was tracking in a direction that made a favourable result unlikely, so continuing would not have rescued it.

The primary endpoint was a composite of death from any cause, non-fatal heart attack or non-fatal stroke at one year [s1]. It occurred in 24.1% of patients in the imaging-guided selective group and 20.7% of patients in the routine invasive group, a hazard ratio of 1.22 with a 95% confidence interval from 0.86 to 1.72 and a p-value of 0.27 [s1].

Two things sit inside those numbers. The confidence interval crosses 1.0 and is wide, so this is not evidence that the selective strategy is worse — the difference is not statistically significant [s1]. But non-inferiority is a higher bar than "no significant difference": to adopt a strategy that does less, you have to show it does not do meaningfully worse, and EVAOLD could not show that. A point estimate of 1.22 pointing the wrong way, in a trial stopped for futility, is the opposite of reassurance.

What the trial does and does not establish

The principal investigator, Professor Gilles Barone-Rochette of the University Hospital of Grenoble, noted that although the selective strategy reduced the need for angiography and its complications, it did not demonstrate non-inferiority for major cardiovascular events [s1]. He added that the results should be read in the context of the early stop and lower-than-expected event rates, and that implementing the stress-imaging strategy proved challenging in this population — a finding in itself [s1].

That last point matters. A strategy that is hard to deliver in frail, older patients is not just statistically inferior; it may be impractical. The trial's honest conclusion is narrow: it does not support swapping routine angiography for a stress-test-first approach in the over-80s, and it says nothing about younger patients, about ST-elevation heart attacks, or about which specific older patients might still reasonably be managed conservatively for other reasons.

EVAOLD was an investigator-led trial funded by the French Ministry of Health through its hospital clinical research programme, not by a device or drug company [s1].

What to watch

The result lands in a run of trials reappraising how aggressively to treat coronary disease, from omitting aspirin around stenting to stopping beta-blockers long after a heart attack. The open question EVAOLD leaves is whether any non-invasive triage tool can safely select older NSTEMI patients for a less invasive path, or whether — given how well tolerated modern angiography has become — the search itself is the wrong one. For now, guideline writers have a clear negative to weigh.

This article describes trial results. It is not medical advice, and nothing here should be used to start, stop or change treatment.

Sources

Sources

  1. EVAOLD trial provides guidance on treating older patients after a heart attack — European Society of Cardiology , August 30, 2026
  2. Evaluation of a Strategy Guided by Imaging Versus Systematic Coronary Angiography (NCT03289728) — ClinicalTrials.gov , April 4, 2018

More on

Related coverage
WHAT THE STUDY ACTUALLY SAYS

Does cardiac rehabilitation after a heart attack work?

Pooled across 85 trials, exercise-based cardiac rehab cut later heart attacks and hospital admissions but barely moved death rates within a year — a real benefit, and a narrower one than the usual pitch.