PCI guided by a novel, artificial-intelligence-powered (AI) method of angiographic fractional flow reserve (FFR) assessment known as AngioFFR, which calculates FFR from a single-view angiogram of the coronary arteries, was noninferior to IVUS for outcomes in patients with significant stenosis, according to new research presented during a Late-Breaking Clinical Trial session at ACC.25 in Chicago and simultaneously published in the The Lancet.

The investigator-initiated, open-label FLAVOUR II trial, conducted between May 2020 and September 2023 at 22 centers across China, randomized 1,839 patients (median age 66 years, 32.1% women) to PCI guided by either AngioFFR (n=923) or IVUS (n=916). Patients were eligible if they had suspected ischemic heart disease and a de novo stenosis ≥50% in a vessel measuring at least 2.5 mm. Among participants, 736 (40%) had a chronic coronary syndrome and 1,087 (59.1%) had an acute coronary syndrome.

Optimal PCI was achieved in 606 (88.9%) patients in the AngioFFR group and 430 (56.5%) in the IVUS group. The proportion of target vessels requiring PCI was significantly lower in the AngioFFR group than the IVUS group, resulting in fewer stent implantations (11.5% reduction) and less frequent use of antiplatelet therapy (5.4% reduction).

Results showed a nonsignificant difference in the primary outcome at 12 months: the composite endpoint of death, myocardial infarction or revascularization occurred in 56 (6.3%) patients in the AngioFFR group and 54 patients (6.0%) in the IVUS group (hazard ratio, 1.04; pnon-inferiority=0.022).

In the two groups, respectively, there were similar rates of mortality (1.8% vs. 1.3%) and recurrent angina (2.8% vs. 3.8%), as well as need for a repeat procedure, and outcomes were similar across subgroups.

"This is the first such study to be conducted in patients with angiographically significant lesions," said Jian'an Wang, MD, PhD, FACC, the study's senior author and editor in chief of JACC: Asia. "Patients whose evaluation was noninvasively guided by the novel, AI-powered technique underwent approximately 10% fewer procedures, and their outcomes were comparable with those for patients whose evaluation was guided by a commonly used ultrasound-based imaging technique."

Investigators will continue to follow patients for at least five years to evaluate whether any differences in long-term effectiveness emerge between the two groups. Among the study limitations was the lack of complex lesions despite them being angiographically significant and being conducted in a single country limiting diversity.