Diagnostic performance of non-contrast whole-heart coronary magnetic resonance angiography combined with black-blood arterial wall imaging in patients with suspected coronary artery disease

Introduction

Whole-heart-coronary-MRA (WH-CMRA) or black-blood-coronary-arterial-wall imaging have been used independently to detect coronary artery disease (CAD) in previous studies [1, 2]. However, the diagnostic performance of a combined approach hasn't been reported yet.

Purpose

To evaluate the diagnostic performance of combined non-contrast WH-CMRA and black-blood-arterial-wall imaging in patients with suspected CAD.

Methods

Thirteen patients with suspected CAD were scanned at 1.5 T (MAGNETOM Sonata, Siemens, Germany) after written informed consent was obtained. WH-CMRA was acquired using a 3D ECG-triggered, navigator-gated, fat-suppressed, T2-prepared steady-state-free-precession sequence. Cross-sectional coronary wall imaging (thinkness = 5 mm) was performed using a 2D-black-blood, navigator and ECG-gated, Turbo-Spin-Echo sequence with asymmetric-adiabatic-spectral inversion-recovery fat suppression [3]. Continous slices without gap for wall imaging were positioned from pre to post of the suspected lesion segment perpendicular to the multi-planar reformats (MPRs) of WH-CMRA. All patients received SCA within 1 week before or after the MR examination. Image quality of WH-CMRA and vessel wall images was evaluated on a segment basis by 4-point scale (1 - poor, 4 - excellent). Segments with a score of 1 in both techniques were excluded from analysis. A positive diagnosis of CAD was made based on the MR images when at least in one of both techniques a stenosis ≥ 50% was detected by visual analysis. Only left main (LM), proximal and middle segments of right coronary artery (RCA)/left anterior descending coronary artery (LAD) were included in our study. CAD was defined positive on SCA when showing a luminal diameter reduction ≥ 50%.

Results

20/65 segments were found stenosis by WH-CMRA. The mean total scan times for WH-CMRA and coronary wall imaging were 13 ± 1.2 min, 2 ± 0.3 min respectively.10 of 20 segments were diagnosed as CAD by SCA. On a segment-based analysis, average image quality of WH-CMRA and arterial wall were 3.5 and 3.45, respectively. Two of 20 segments of wall image score were 1 because of rapid heart rate. MR results did not agree with SCA in one of 20 caused by high BMI (34.17) and irregular breathing. The sensitivity of WH-CMRA only and WH-CMRA combined with arterial wall were (8/10) and (9/10), NPV (10/12) and (10/11), PPV(8/8) and (9/9) respectively. No difference of specificity (10/10) between the two groups Figure 1 and table 1.

Table 1 Patient population and results of non-contrast WH-CMRA and black blood arterial wall compared with SCA.

Figure 1

59-Y man w. typical chest pain. MPRs of WH-CMRA (1b, 1c) show significant stenosis (≥ 50%) in middle segments of LAD, consistent with findings of SCA (1a). 2D black blook coronary wall images were acquired pre/mid/post LAD lesion (1d). Mid slice shows stenosis >50% while lumen in slice pre, post is <50%.

Conclusion

The combination of WH-CMRA and black blood coronary wall imaging improves diagnostic accuracy to detect CAD over WH-CMRA alone. The study with larger sample size is under investigation.