Combined non-contrast enhancement 3D bright-blood and 2D black-blood coronary magnetic resonance angiography for the assessment of coronary artery stenosis

Background: Non-contrast coronary magnetic resonance angiography (CMRA) offers a radiation- and contrast-free alternative for evaluating coronary artery disease, yet its clinical adoption has been constrained by suboptimal specificity, primarily due to flow-related artifacts and signal loss inherent to bright-blood imaging. Black-blood vessel wall imaging, by suppressing intraluminal blood signal, enables direct visualization of the coronary wall and may complement bright-blood CMRA to improve diagnostic accuracy. However, the additive value of a targeted sequential protocol combining these two techniques remains to be established. The objective of this study was to evaluate whether a sequential protocol combining three-dimensional (3D) bright-blood CMRA with lesion-directed two-dimensional (2D) black-blood coronary magnetic resonance (CMR) improves the diagnostic performance of non-contrast CMRA for detecting coronary artery stenosis, using computed tomography angiography as the reference standard.

Methods: From January 2025 to April 2025, consecutive patients with suspected coronary artery disease who were scheduled for dual-source computed tomography angiography and consented to undergo 3.0-T non-contrast CMRA were enrolled. The imaging protocol comprised free-breathing 3D bright-blood whole-heart CMRA, followed by lesion-directed 2D black-blood CMRA performed exclusively in segments with equivocal or suspicious findings on bright-blood imaging. Coronary computed tomography angiography (CCTA) served as the primary reference standard, with ICA performed in a subset of nine patients. Diagnostic performance for detecting significant (>50% diameter stenosis) and mild (25–49%) stenosis was assessed on a per-segment basis. Comparative analyses between bright-blood CMRA alone and the combined approach were performed using the McNemar test for paired proportions and DeLong’s method for receiver operating characteristic (ROC) curves.

Results: A total of 114 vessel segments from 104 patients (mean age 63±13 years; 63 men)were included in the final analysis. The combination of bright-blood CMRA and 2D black-blood CMR demonstrated superior diagnostic performance compared with bright-blood CMRA alone for identifying significant stenosis (>50% diameter stenosis). Sensitivity was 97.9% [95% confidence interval (CI), 88.9–99.9%] versus 87.5% (95% CI, 72.3–93.9%; P=0.031), specificity was 98.5% (95% CI, 91.8–100.0%) versus 68.1% (95% CI, 52.4–76.5%; P<0.001), accuracy was 98.2% (95% CI, 93.8–99.8%) versus 76.3% (95% CI, 64.6–81.5%; P<0.001), and the area under the receiver operating characteristic curve (AUC) was 0.98 (95% CI, 95.2–99.1%) versus 0.75 (95% CI, 66.2–82.1%; P<0.001). In addition, the combined approach correctly identified non-significant stenosis (25–49% diameter stenosis) in 32 of 35 segments and significantly improved sensitivity [91.4% (95% CI, 76.8–98.0%) vs. 71.4% (95% CI, 53.7–85.4%); P=0.039].

Conclusions: The combination of 3D bright-blood CMRA and 2D black-blood CMR significantly enhances the diagnostic performance of non-contrast CMRA for detecting both significant and mild coronary artery stenosis.