Background Coronary artery bypass grafting (CABG) targets are selected by angiographic stenosis, whereas percutaneous coronary intervention increasingly relies on functional assessment. Whether quantitative flow ratio (QFR), a wire-free angiography-derived index, relates to graft outcomes is unclear. We examined preoperative QFR and 1-year graft patency. Methods We retrospectively studied 491 patients undergoing isolated CABG at two heart centers, 2006–2019. Investigators blinded to bypass design and outcome measured preoperative QFR in all analyzable major vessels. Protocol angiography assessed 1-year graft status. Generalized estimating equations accounting for within-patient clustering related QFR (per 0.1-unit increment) to graft occlusion, unadjusted and adjusted for target vessel, graft type, age, sex, diabetes, and ejection fraction. Linearity and effect modification by graft type and target territory were assessed. Results Median QFR among 1161 analyzable vessels was 0.66 (interquartile range, 0.48–0.80); among 1036 bypassed vessels, higher QFR was associated with greater 1-year graft occlusion risk (unadjusted odds ratio [OR], 1.27; 95% confidence interval [CI], 1.15–1.42; adjusted OR, 1.15; 95% CI, 1.04–1.27; p = 0.008). The association was linear (standardized occlusion probability, 8.2% at QFR 0.50; 13.1% at 0.90). Effect modification by target territory was borderline (joint interaction p = 0.025; Holm-adjusted p = 0.050); this exploratory finding requires confirmation. Graft type did not modify the association (p = 0.28). Conclusions Higher preoperative QFR, indicating functionally less significant stenosis, was associated with greater 1-year graft occlusion risk. This hypothesis-generating, group-level association establishes neither causation nor predictive utility for individual grafts. Prospective validation is needed.
Background Angina with no obstructive coronary arteries is often due to coronary vascular dysfunction, which is increasingly described as a contributor and predictor of heart failure with preserved ejection fraction. We tested the hypothesis that coronary vascular dysfunction contributes to cardiac dysfunction. Methods Sixty-eight participants with suspected coronary vascular dysfunction underwent LV pressure-volume assessment at rest and during isometric handgrip stress. Coronary function testing was performed by infusing standard doses of adenosine, acetylcholine, and nitroglycerine through a guiding catheter placed in the left main coronary artery, per established protocols. Results Of the participants studied, 12 had normal coronary vascular function while 56 had coronary vascular dysfunction. At rest, there were no differences between groups in any invasive pressure or volumetric measurement. With isometric handgrip, peak filling rate increased significantly in participants with normal coronary vascular function (0.56 ± 0.31 vs. 0.85 ± 0.68 1/s/mmHg) but remained unchanged in participants with coronary vascular dysfunction (0.51 ± 0.22 vs. 0.55 ± 0.43 1/s/mmHg; group x time interaction, p = 0.04). Moreover, left ventricular end-diastolic pressure decreased significantly with isometric handgrip in participants with normal coronary vascular function (12 ± 3 vs. 9 ± 4 mmHg) but increased in participants with coronary vascular dysfunction (14 ± 4 vs. 15 ± 6 mmHg; group x time interaction, p = 0.03). Conclusions Coronary vascular dysfunction was associated with an impaired ability of the LV to adapt its filling during stress. Further work is needed to assess potential clinical significance.
