#angiography-based-physiology #fractional-flow-reserve #coronary-physiology #intermediate-coronary-lesions #pci-guidance

Angiography-Based Coronary Physiology

Definition

Angiography-based coronary physiological indices derive a functional assessment of coronary stenosis severity (analogous to fractional flow reserve) directly from coronary angiographic images, using 3D quantitative coronary angiography (QCA) and computational fluid dynamics models, without requiring a pressure wire or pharmacologic hyperemia. The examples validated in randomized outcome trials are vessel fractional flow reserve (vFFR, FAST III), FFRangio (ALL-RISE) and, in STEMI, functional coronary angiography (AIR-STEMI); quantitative flow ratio (QFR) failed noninferiority in FAVOR III Europe. These tools aim to make physiological guidance universally accessible, shorten procedures, and eliminate wire-related and adenosine-related complications.

Key Concepts

How vFFR Works

vFFR (Pie Medical Imaging) reconstructs coronary geometry in three dimensions from two angiographic projections acquired ≥30° apart, after intracoronary nitrate administration
Aortic-root pressure is measured invasively via the guiding catheter tip (not eliminated — only the intracoronary wire is removed)
Blood flow behaviour is modelled using physical laws: viscous resistance and separation loss effects (Gould and Kirkeeide formulations from the 1980s)
Output: a single vFFR value analogous to wire-measured FFR; ≤0.80 = hemodynamically significant, >0.80 = defer revascularization
Requires local analyst certification; performed directly at the cath lab workstation

Diagnostic Accuracy vs Pressure-Wire FFR

Prospective multicenter validation studies (FAST 2020, FAST II 2022, FAST EXTEND 2021) demonstrated diagnostic accuracy of vFFR comparable to pressure-wire FFR for identifying hemodynamically significant stenoses (sources/vffr-fastiii-nejm-2026, rating: very high)
Other angiography-based indices include quantitative flow ratio (QFR), AngioFFR, and caFFR — each use different computational approaches and validation datasets

FAST III — First Head-to-Head RCT: vFFR vs Wire FFR

Design: 2,235 patients with intermediate coronary-artery lesions (30–80% stenosis); chronic or acute coronary syndrome; 37 European centers; 1:1 vFFR vs FFR revascularization guidance; 1-year follow-up (sources/vffr-fastiii-nejm-2026, rating: very high)
Primary endpoint (death/MI/revascularization at 1 year): 7.5% (vFFR) vs 7.5% (FFR); risk difference −0.02 pp (95% CI −2.25 to 2.21); P=0.004 for noninferiority — vFFR noninferior to FFR
Study-vessel failure: 4.0% vs 4.6% (NS)
Death: 2.2% vs 2.3% (NS); MI: 2.9% vs 2.4% (NS); Any revascularization: 4.2% vs 4.2%
Quality of life (SAQ score at 1 year): 87.3 vs 86.8 — equivalent

FFRangio — ALL-RISE (NEJM 2026)

FFRangio (CathWorks) reconstructs a 3D coronary tree from three angiographic images and applies a resistance model to estimate pressure gradients from lumen contours; it requires no pressure wire and no hyperaemic agent, and uses the same ≤0.80 threshold as wire FFR (sources/ffrangio-allrise-nejm-2026, rating: very high)
ALL-RISE randomized 1,930 patients with intermediate stenoses (50–90%) 1:1 to FFRangio or pressure-wire assessment (FFR in 55.5%, non-hyperaemic pressure ratio in 44.5%) at 59 international sites (sources/ffrangio-allrise-nejm-2026, rating: very high)
Primary endpoint (death, MI, or unplanned clinically indicated revascularization at 1 year): 6.9% vs 7.1%; HR 0.98 (95% CI 0.70–1.39); difference −0.2 percentage points; upper boundary of the one-sided 97.5% CI 2.1 points, below the 3.5-point margin; P<0.001 for noninferiority (sources/ffrangio-allrise-nejm-2026, rating: very high)
FFRangio classified more lesions as significant (43.0% vs 37.0%) and led to more PCI (44.3% vs 35.4% of lesions; OR 1.45; 95% CI 1.24–1.70) with equivalent clinical outcomes — the same "over-detection" pattern seen with vFFR (sources/ffrangio-allrise-nejm-2026, rating: very high)
Workflow advantages: assessment time 6 vs 8 min; procedural time 39 vs 42 min; fluoroscopy 9 vs 12 min; contrast 100 vs 105 mL (all statistically significant) (sources/ffrangio-allrise-nejm-2026, rating: very high)
Safety: BARC 3/4/5 bleeding 0.7% vs 1.5%; no AKI in the FFRangio arm; periprocedural angiographic complications 1.0% vs 0.5% (NS) (sources/ffrangio-allrise-nejm-2026, rating: very high)
Subgroups suggested a vessel interaction — lower events with FFRangio in LAD lesions and higher in non-LAD lesions — and a possible difference according to whether the comparator used FFR or a non-hyperaemic pressure ratio; the trial was underpowered for these analyses (sources/ffrangio-allrise-nejm-2026, rating: very high)
ALL-RISE differs from FAVOR III Europe: QFR was not noninferior to FFR, whereas FFRangio was. The authors attribute this to different computational methods and to greater operator interaction and inter-observer variability with QFR (the REPEAT-QFR substudy) (sources/ffrangio-allrise-nejm-2026, rating: very high)

Key Discordance: vFFR Detects More Functionally Significant Lesions

Functionally significant lesions (≤0.80): 40.9% (vFFR) vs 31.3% (FFR) — a 9 percentage point gap (sources/vffr-fastiii-nejm-2026, rating: very high)
This resulted in more revascularizations with vFFR (45.0% vs 36.0%) despite equivalent outcomes
Mechanistic explanation: angiography-based tools assume flow based on vessel geometry (reference diameter) rather than measuring actual hyperemic flow; true microvascular resistance, demand, and patient-specific hyperemia response are unaccounted for → computed pressure gradient exceeds true gradient → lower vFFR value → higher lesion detection rate
Longer or tapered lesions may be further overestimated due to accumulated resistance across diffuse disease
This vFFR-FFR discordance is a class effect — parallel findings in FAVOR III Europe (QFR vs FFR) showed QFR also detected more lesions, albeit with different clinical outcome implications

Practical Advantages Over Wire-Based FFR

No intracoronary wire needed → eliminates wire-related complications (dissection, perforation, spasm, wire bias artefact)
No pharmacologic hyperemia → eliminates adenosine-related effects (AV block, bronchospasm, discomfort, cost)
Procedure time shorter by ~5 minutes (55.8 vs 60.9 min; diff −5.13 min; 95% CI −8.55 to −1.71) (sources/vffr-fastiii-nejm-2026, rating: very high)
Fewer intraprocedural complications at time of physiological assessment: 3.7% vs 6.0% (sources/vffr-fastiii-nejm-2026, rating: very high)
Enables retrospective physiological analysis of angiographic images already acquired

Context: Where Does vFFR Fit Relative to Other Physiology Tools?

FFR (wire + adenosine): Gold standard; highest evidence base; requires additional equipment; adenosine complications
iFR (wire, no adenosine): Validated equivalent to FFR in stable CAD; ≤0.89 threshold; also wire-based but no adenosine; STEMI false-positive risk (~11%)
vFFR (no wire, no adenosine): Now validated noninferior to FFR in a large RCT for clinical outcomes; higher lesion detection rate; shorter procedure; lower complication rate
QFR (no wire, no adenosine): Class IB recommendation (ESC 2024 CCS guidelines) after FAVOR III China (superior to angiography); FAVOR III Europe showed higher MACE than FFR-guided revascularization (different result from FAST III)
The divergence between FAVOR III Europe (QFR vs FFR) and FAST III (vFFR vs FFR) highlights that not all angiography-based indices are equivalent

FAVOR III Europe vs FAST III — Why Divergent Results?

FAVOR III Europe (Lancet 2024; n=2,000): QFR-guided revascularization showed higher MACE than FFR-guided revascularization — QFR performed worse than FFR
FAST III (NEJM 2026; n=2,235): vFFR-guided revascularization showed equivalent MACE to FFR-guided revascularization — vFFR noninferior to FFR
Both trials used the same ≤0.80 threshold; the difference in outcome may reflect computational methodology differences between QFR and vFFR, or differences in patient populations and center characteristics
The exact mechanism of QFR inferiority vs FFR in Europe remains unexplained; vFFR may have a more accurate flow model

Contradictions / Open Questions

Higher revascularization rate with vFFR without better outcomes: vFFR drove 9 pp more revascularizations (45% vs 36%) with identical outcomes. Does this indicate that vFFR better identifies prognostically relevant lesions (revascularization beneficial but not captured in 1-year composite), or that vFFR over-detects, adding stents without benefit? A prespecified cost-effectiveness analysis is planned (sources/vffr-fastiii-nejm-2026, rating: very high)
vFFR in ACS — diagnostic reliability gap now partly closed by outcome data: only 18.7% of FAST III patients had ACS, and vFFR reliability in ACS was considered less established (post-ACS microvascular dysfunction may magnify vFFR–FFR discordance). AIR-STEMI (NEJM 2026) addresses the outcome side of that gap: in 1,823 STEMI patients with multivessel disease, complete revascularization guided by functional coronary angiography (angiography-derived distal FFR, FFR ≤0.80 treated, core-lab analysed, no pressure wire or hyperaemic agent) beat angiography-guided completion — primary composite of death/MI/stroke-TIA/ischaemia-driven revascularization 8.9% vs 13.7% (HR 0.62; 0.47–0.83), with a better safety composite of contrast-associated AKI or major bleeding (4.6% vs 7.1%; HR 0.63). This is the first randomized outcome evidence for angiography-derived physiology, as distinct from FAST III's diagnostic non-inferiority evidence (sources/AIR-STEMI-NEJM-2026, rating: very high)
- Note the contrast with FAST III's revascularization signal: vFFR treated more lesions than wire FFR (45% vs 36%) with identical outcomes, whereas in AIR-STEMI physiology-guided treatment of fewer vessels than angiography produced better outcomes. The two are compatible — angiography over-treats most, and vFFR sits between angiography and wire FFR — but they mean "angiography-derived physiology" is not a single behaviour, and its calibration relative to both comparators remains unsettled
- AIR-STEMI used the derived physiology for two purposes (lesion selection and PCI planning via pullback/stent sizing); which drives the benefit is unknown
Long-term outcomes unknown: The higher stent burden with vFFR (more lesions treated) may affect long-term stent thrombosis, restenosis, and repeat revascularization beyond 1 year
QFR vs vFFR vs FFRangio — three different answers: FAVOR III Europe found QFR not noninferior to FFR, FAST III found vFFR noninferior, and ALL-RISE found FFRangio noninferior. The divergence is attributed to computational method and operator dependence, but no trial compares the angiography-based indices head to head — so "angiography-derived physiology" is not a single entity (sources/ffrangio-allrise-nejm-2026, rating: very high)
More lesions treated without better outcomes — again. FFRangio classified 43.0% of lesions as significant versus 37.0% with pressure-wire assessment and drove PCI in 44.3% versus 35.4% of lesions, yet 1-year outcomes were identical. Whether this is appropriate treatment of grey-zone lesions or over-stenting is unresolved; ALL-RISE was not powered for longer-term stent outcomes (sources/ffrangio-allrise-nejm-2026, rating: very high)
Industry sponsorship in ALL-RISE: CathWorks funded the trial and participated in site selection, data collection and monitoring, albeit with independent statistical analysis by the Cardiovascular Research Foundation — a conflict that applies to the whole angiography-derived physiology literature (sources/ffrangio-allrise-nejm-2026, rating: very high)
Generalizability of FFRangio: the trial required anatomy amenable to both pressure-wire and FFRangio assessment and mostly enrolled stable patients without biomarker elevation; image quality requirements (three projections ≥30° apart, no table movement) may not be met in all laboratories (sources/ffrangio-allrise-nejm-2026, rating: very high)
Generalizability beyond high-volume European centers: vFFR requires specific training, certification, and accurate angiographic acquisition technique; adoption in lower-volume centers may yield different results

Connections

Related to concepts/Fractional-Flow-Reserve — wire-based gold standard comparator; ≤0.80 threshold shared
Related to concepts/Instantaneous-Wave-Free-Ratio — wire-based resting alternative to FFR; no adenosine
Related to concepts/Multivessel-PCI-STEMI-Timing — physiological guidance of nonculprit lesions in ACS context
Related to concepts/NSTEMI-Elderly-Invasive-Strategy — invasive physiology guidance in ACS populations
Related to concepts/Complete-Revascularization — physiology-based selection of nonculprit lesions