High-resolution non-contrast free-breathing coronary cardiovascular magnetic resonance angiography for detection of coronary artery disease: validation against invasive coronary angiography

Abstract

Background Coronary artery disease (CAD) is the single most common cause of death worldwide. Recent technological developments with coronary cardiovascular magnetic resonance angiography (CCMRA) allow high-resolution free-breathing imaging of the coronary arteries at submillimeter resolution without contrast in a predictable scan time of ~ 10 min. The objective of this study was to determine the diagnostic accuracy of high-resolution CCMRA for CAD detection against the gold standard of invasive coronary angiography (ICA). Methods Forty-five patients (15 female, 62 ± 10 years) with suspected CAD underwent sub-millimeter-resolution (0.6 mm3) non-contrast CCMRA at 1.5T in this prospective clinical study from 2019–2020. Prior to CCMR, patients were given an intravenous beta blockers to optimize heart rate control and sublingual glyceryl trinitrate to promote coronary vasodilation. Obstructive CAD was defined by lesions with ≥ 50% stenosis by quantitative coronary angiography on ICA. Results The mean duration of image acquisition was 10.4 ± 2.1 min. On a per patient analysis, the sensitivity, specificity, positive predictive value and negative predictive value (95% confidence intervals) were 95% (75–100), 54% (36–71), 60% (42–75) and 93% (70–100), respectively. On a per vessel analysis the sensitivity, specificity, positive predictive value and negative predictive value (95% confidence intervals) were 80% (63–91), 83% (77–88), 49% (36–63) and 95% (90–98), respectively. Conclusion As an important step towards clinical translation, we demonstrated a good diagnostic accuracy for CAD detection using high-resolution CCMRA, with high sensitivity and negative predictive value. The positive predictive value is moderate, and combination with CMR stress perfusion may improve the diagnostic accuracy. Future multicenter evaluation is now required.

Deep Learning‐Based Acceleration of Compressed Sensing for Noncontrast‐Enhanced Coronary Magnetic Resonance Angiography in Patients With Suspected Coronary Artery Disease

“…Among all coronary segments with a luminal diameter ≥1.5 mm on the CTA images, a majority of the coronary segments on MRA were diagnostic. Furthermore, coronary MRA with CSAI in our study achieved satisfactory diagnostic accuracy within a shorter scan time comparable to previous noncontrast‐enhanced coronary MRA studies with 3D patch‐based low‐rank reconstruction 5,37 . The high specificity and NPV suggested that CSAI‐based coronary MRA might have a potential role as a rule‐out test for CAD.…”

“…Furthermore, coronary MRA with CSAI in our study achieved satisfactory diagnostic accuracy within a shorter scan time comparable to previous noncontrast-enhanced coronary MRA studies with 3D patchbased low-rank reconstruction. 5,37 The high specificity and NPV suggested that CSAI-based coronary MRA might have a potential role as a rule-out test for CAD. The moderate PPV might primarily be due to the presence of many moderate (25%-49%) stenotic segments, overestimated by observers as ≥50% lesions.…”

Feasibility of 3.0 T balanced fast field echo non-contrast-enhanced whole-heart coronary magnetic resonance angiography

“…Only segments with the score of 3 or 4 assessed by two radiologists consistently were included in the final stenosis assessment compared with CCTA. We incorporated 8 coronary segments into the analysis refer to the classification of the American Heart Association ( 14 ), as follows: (segment 1) RCA proximal, (segment 2) RCA middle, (segment 3) RCA distal, (segment 6) LAD proximal, (segment 7) LAD middle, (segment 8) LAD distal, (segment 11) LCX proximal, and (segment 13) LCX distal. The third radiologist evaluated the image quality when two observers assessed differently in the axial orientation.…”

Diagnostic accuracy of whole heart coronary magnetic resonance angiography: a systematic review and meta-analysis

“…The publication years of these studies ranged from 2005 to 2022; 23 of these studies utilized 1.5 T MRI technology [ 13 – 25 , 28 , 30 , 31 , 33 – 37 , 44 – 46 ], while 11 utilized 3.0 T technology [ 26 , 27 , 29 , 32 , 38 – 43 , 45 ]. The countries of publication were diverse, with China having 11 reports [ 18 , 26 , 32 , 34 , 38 , 39 , 41 – 45 ], Germany having 8 reports [ 13 , 15 , 19 , 25 , 27 , 29 , 31 , 35 ], Japan having 7 reports [ 14 , 17 , 24 , 28 , 30 , 37 , 40 ], United Kingdom [ 21 , 46 ] having two reports and various other countries such as the United States [ 20 ], Korea [ 16 ], Switzerland [ 36 ], Portugal [ 33 ], Belgium [ 23 ], and Turkey [ 22 ] having one report each. In terms of study design, one study employed a prospective multicenter design [ 28 ], while 19 employed prospective single-center designs [ 13 – 15 , 17 , 21 , 23 , 25 – 27 , 29 – 32 , 37 – 39 , 43 – 46 ].…”