ALL‑RISE: AI‑Supported Coronary Flow Assessment Performs Similarly to Gold‑Standard Wire‑Based Testing

The ALL‑RISE trial, presented at ACC.26 and simultaneously published in the New England Journal of Medicine, shows that FFRangio, an AI‑supported, angiography‑derived method for fractional flow reserve (FFR), performs similarly to traditional wire‑based invasive FFR in patients with coronary artery disease (CAD) referred for PCI. 

Investigators randomly assigned patients undergoing coronary angiography who were found to have at least one intermediate coronary stenosis to physiological assessment with measurements derived from angiographic images (FFRangio) or with pressure-wire–based measurements.

In this large, international noninferiority study, 1,930 patients were randomized 1:1 to FFRangio or conventional pressure‑wire‑based FFR, with a mean age of 68, 25% women, and high rates of prior MI, PCI, hypertension, hyperlipidemia, and diabetes.

At one year, the composite primary endpoint of death, MI, or unplanned clinically indicated revascularization occurred in 6.9% of the FFRangio group versus 7.1% of the wire‑based group (HR 0.98), meeting the prespecified criterion for noninferiority. 

There were no differences in bleeding, acute kidney injury, or procedure‑related adverse events. The FFRangio approach was also faster (39 vs. 42 minutes per case), required less fluoroscopy, less contrast, and avoided additional coronary manipulation beyond standard angiography.

The senior investigator notes that FFRangio provides similar clinical outcomes at one year to the current gold‑standard wire‑based assessment, while being less invasive. 

In an accompanying editorial, an expert stresses that the broader implication is a shift away from angiographic anatomy alone toward routine, integrated physiologic assessment; AI‑derived FFR may become a practical, first‑line tool during diagnostic coronary angiography to guide revascularization decisions. 

Limitations include the trial’s open‑label design and exclusion of patients with prior CABG, but the data support wider adoption of physiology‑guided PCI using software‑based tools like FFRangio.