Impact of coronary lesion location on the diagnostic performance of Murray law-based quantitative flow ratio
W Zuo, Z Ji, R Sun, X Zhang, G Ma, R Yang, H BaoAbstract
Background
Coronary physiology-guided revascularisation improves outcomes, yet invasive fractional flow reserve (FFR) is underused due to complexity and cost. Angiography-based FFR, including Murray law-based QFR (μQFR), allows wire-free assessment without hyperaemia. Whether its diagnostic accuracy varies by coronary stenosis location remains unclear.
Purpose
To assess μQFR accuracy versus FFR across coronary arteries and lesion segments, and to determine the influence of lesion location.
Methods
Consecutive patients undergoing coronary angiography with pressure wire-based FFR assessment for at least one intermediate stenosis (30–70% by visual estimation) were analysed. Vessels with left main disease, recent myocardial infarction, in-stent restenosis, or inadequate image quality were excluded. FFR ≤0.80 defined haemodynamic significance. μQFR was computed offline by blinded analysts from single angiographic projections. Per-vessel correlation and agreement with FFR were assessed by Spearman and Bland–Altman analyses. ROC curves were compared with QCA-derived diameter stenosis (QCA-%DS) using the DeLong test.
Results
497 vessels from 460 patients were analysed. Haemodynamic significance occurred in 107/335 LAD (32%), 13/79 LCX (16.5%), and 10/83 RCA (12.0%) vessels, and in 66/245 proximal (27.0%), 58/222 middle (26.1%), and 6/30 distal (20.0%) segments. μQFR correlated well with invasive FFR across different arteries (LAD: rho=0.770; LCX: rho=0.759; RCA: rho=0.684; P<0.001 for all) and segments (proximal: rho=0.787; middle: rho=0.761; distal: rho=0.820; P<0.001 for all) (Figure 1), with minimal bias (LAD: 0.010±0.052; LCX: 0.001±0.045; RCA: -0.005±0.061; proximal: 0.008±0.054; middle: 0.004±0.051; distal: 0.004±0.051). Per-vessel analysis showed that μQFR had higher diagnostic accuracy than QCA-%DS in LAD 85.4% (95% confidence interval, 81.1–89.0) vs. 71.6% (66.5–76.4, P<0.001), proximal 86.5% (81.6–90.5) vs. 73.9% (67.9–79.3, P<0.001), and middle segments 86.9% (81.8–91.1) vs. 75.2% (69.0–80.8, P<0.001), a trend toward improvement in LCX 88.6% (79.5–94.7) vs. 84.8% (75.0–91.9, P=0.051), while differences in RCA 89.2% (80.4–94.9) vs. 81.9% (72.0–89.5, P=0.396) and distal segments 83.3% (65.3–94.4) vs. 90.0% (74.5–97.9, P=0.266) were not statistically significant (Figure 2).
Conclusions
μQFR provides stable, location-independent diagnostic performance across coronary arteries and lesion segments. It outperforms angiographic diameter stenosis assessment in commonly affected vessels and segments, while maintaining reliable accuracy in distal lesions, supporting its use as a consistent, wire-free tool for physiological lesion assessment in routine practice.Figure 1For image description, please refer to the figure legend and surrounding text.Figure 2For image description, please refer to the figure legend and surrounding text.