Functional Coronary Angiography to Indicate and Guide Revascularization in STEMI Patients With Multivessel Disease (AIR-STEMI)

doi:10.1056/nejmoa2605373

AIR-STEMI

Patient / Population Intervention / Exposure Comparison Outcome

In 1,823 patients with STEMI and multivessel disease, angiography-derived FFR-guided non-culprit PCI reduced death, stroke, reinfarction or ischaemia-driven revascularisation compared with angiography-guided PCI.

N

1,823patients

Design

Prospective, international, multicentre, randomised, open-label trial with blinded adjudication of outcomes by an independent clinical events committee; 1:1 randomisation after successful culprit-lesion PCI; non-culprit PCI indicated and planned by angiography-derived FFR versus conventional angiography; follow-up censored when the last enrolled patient reaches 1 year (average about 18 months), with endpoints also assessed at 3 and 5 years

Endpoint

Patient-oriented composite of all-cause death, cerebrovascular accident, reinfarction or ischaemia-driven revascularisation

Relevance

2Important — one of several pillars.

ResultDeath, stroke, reinfarction or ischaemia-driven revascularisation 8.9% vs 13.7% over a median 17.9 months (HR 0.62, 95% CI 0.47-0.83; P < 0.001). Cardiovascular death or myocardial infarction HR 0.52 (0.37-0.73); myocardial infarction 3.8% vs 8.0% (HR 0.47), both spontaneous and procedure-related. Contrast-associated kidney injury 2.2% vs 4.0% (HR 0.55). The physiology arm used 38% fewer staged procedures, treated 45% fewer non-culprit vessels and implanted 31% less stent length.

Biscaglia S, Campo G, et al. N Engl J Med. 2026; published online 29 August. 10.1056/nejmoa2605373

Discussion & critique

Reading physiology off the angiogram rather than crossing every lesion with a wire cut the composite by nearly two fifths, and the mechanism is visible in the procedure log: fewer staged procedures, fewer vessels treated, less stent, less contrast. Reinfarction fell for both spontaneous and procedure-related events, which argues that some of the benefit is simply harm not done. Two caveats: the trial is open-label with ischaemia-driven revascularisation in the primary endpoint, a component the operator can influence, and the FFR values came from a central core laboratory rather than the local cath lab.