Application Of 3.0-Tesla MR In Coronary MR Angiography And The Detection Of Coronary Artery Microembolization

Part IDiagnostic Performance of Contrast-enhanced Whole-heart Coronary Magnetic Resonance Angiography at 3.0T in Patients with Suspected CADBackground:Contrast-enhanced whole-heart magnetic resonance angiography (MRA) at 3.0-Tesla has been performed in previous studies. However, further research about its diagnostic value in CAD patients is still needed for inconsistent conclusions.Purpose:To prospectively evaluate contrast-enhanced whole-heart magnetic resonance angiography (MRA) at 3.0-Tesla in diagnoses of patients with suspected cardiac artery disease (CAD), and to evaluate whether contrast-enhanced whole-heart MRA can characterize myocardial infarction with patterns similar to those obtained by conventional 2D phase-sensitive inversion-recovery (PSIR) TrueFISP sequence.Material and methods:The contrast-enhanced whole-heart MRA was performed in consecutive 70 patients with suspected CAD by using a 3.0-T MR system (MAGNETOM Verio, Siemens AG Healthcare, Erlangen, Germany). Three dimensional MR angiographic data was acquired with a respiratory-gated, electrocardiography-triggered, inversion-recovery, segmented 3D-FLASH sequence. A 0.2mmol/kg body weight of Gd-BOPTA was slowly injected using a power injection at a rate of 0.3ml/s, followed by a chaser of 20ml saline at same rate. Data acquisition began 60 seconds after the initialization of contrast administration. At last, 2D segmented, breath-hold PSIR TrueFISP delayed enhancement imaging sequence was performed in left ventricular short-axis view,2-chamber and 4-chamber view (TI=300ms). Diagnostic accuracy of MRA in detecting significant stenosis (≥50%) was evaluated using quantitative coronary angiography (QCA) as the reference standard. The MRA reconstructed images and PSIR TrueFISP images were compared using Kappa test for detecting of myocardial infarction.Results:Coronary MRA was successfully performed on 58 patients, and was interrupted in 12 patients for angina pectoris, allergic reactions, motion and arrhythmia. Five patients were aborted for poor imaging quality. The acquisition time was (620±176)s, ranging from 392 to 1095s. The Mean heart rate was 64.9±11.2 beats/min and mean trigger-delay time was 597ms. Coronary MRA was acquired during diastole in 44 patients (with mean heart rate 63.9±10.7 beats/min) and during systole in 9 patients (with mean heart rate 75.0±9.2 beats/min).The average navigator efficiency was 35%. In 53 patients,499 segments were assessable in MRA and main reasons for nonassessable segments were motion artifacts, total occlusion in proximal segments and small diameter of coronary artery. The whole-heart MRA correctly identified significant CAD in 42 of 44 patients with sensitivity 95.5% and accuracy 94.3%. Based on per-vessel, the sensitivity, specificity, negative predictive value, positive predictive value and accuracy were 86.8%,93.3%,92.6%,88.0% and 91.0% respectively. Based on per-segment level, these values were 83.6%,95.8%,96.0%, 82.8% and 93.4% respectively. Myocardial infarction was observed in 12 of 53 patients and 64 segments by PSIR TrueFISP images. And 60 segments in the same 12 patients were diagnosed with myocardial infarction by MRA images. The consistency of these two imaging sequence in diagnosis of myocardial infarction was excellent with Kappa value 0.91.Conclusions:The contrast-enhanced whole-heart MRA at 3.0-T had high accuracy in diagnoses of significant stenosis (≥50%) in patients with suspected CAD. Myocardial infarction could be accurately detected from 3D-FLASH MRA images with high consistency with referenced standard 2D PSIR TrueFISP sequence. It is feasible to obtain information about coronary artery stenoses and myocardial viability in a single contrast-enhanced whole-heart 3.0-T MRA examination, which may be helpful to clinical diagnose and treatment. PartⅡWhole-heart coronary artery magnetic resonance angiography:a comparison study between 3.0T contrast-enhanced FLASH and 1.5T-SSFP sequenceBackground:Steady-state free precession (SSFP) sequence has gained successfully employment in MRA at 1.5T for its excellent signal to noise ratio (SNR) and contrast to noise ratio (CNR).Fast low angle shot (FLASH) sequence is less sensitive to static field inhomogeneous and is more feasible at 3.0T compared with SSFP.Purpose:To compare the CNR, SNR and image quality of contrast-enhanced whole-heart coronary MRA at 3.0T and noncontrast SSFP at 1.5T in patients with suspected coronary artery disease (CAD).Materials and methods:Thirty patients were enrolled. Fifteen patients underwent coronary MRA by using respiratory-gated, electrocardiography-triggered, T2-prepared, segmented 3D-TrueFISP sequence without administration of contrast medium at 1.5T (Magnetom Avanto, Siemens AG Erlangen, Germany). Fifteen patients (with age, gender, body weight and heart rate matched with patients at 1.5T) underwent cardiac MRA using respiratory-gated, electrocardiography-triggered, inversion-recovery, segmented 3D-FLASH sequence with slow infusion of Gd-BOPTA (0.2mmol/kg body weight at a rate of 0.3ml/s) at 3.0T (MAGNETOM Verio, Siemens AG Healthcare, Erlangen, Germany). SNR and CNR were calculated from the original images. Region of interest for measure the blood signal intesnsity was placed in the ascending aorta at the level of left coronary sinus. Image quality was evaluated based on motion artifact, coronary artery marginal sharpness and background suppression. The results of the measurements were statistically analyzed by independent sample T-test and nonparametric Wilcoxon signed rank test, and P value<0.05 indicates significant difference.Results:The average SNR, CNR and image quality score was 13.9±2.3/11.2±2.1, 10.9±2.6/9.2±2.9 and 3.0±1.0/2.9±1.0 at 1.5T and 3.0T respectively. The overall SNR at1.5T was significantly higher than that at 3.0T (13.9±2.3 vs 11.2±2.1, P<0.05), whereas the CNR and image quality had no significantly difference at 1.5T and 3.0T.Conclusions:Whole-heart coronary MRA in SSFP sequence at 1.5T demonstrated higher SNR, similar CNR and image quality compared with contrast-enhanced whole-heart coronary MRA in FLASH sequence at 3.0T. Greater sample size and self-control method are still needed for further research in future. PartⅢ3.0-Tesla MR Imaging Characterization after Coronary Artery Microembolization in SwineBackground:Despite complete reperfusion of epicardial culprit artery, progressive flow impairment and tissue hypoperfusion persists within more than 25% of infarct area for coronary microembolization (CME), which gradually giving rise to left ventricular (LV) remodeling, LV dysfunction and cardiovascular events. Magnetic resonance (MR) imaging has the ability to help noninvasively measure myocardial perfusion, infarction size and LV function. However, to date, few studies have addressed myocardial imaging characterization after CME by using 3.0-Tesla MR and further research is needed.Purpose:The aim of this study was to use 3.0T cardiac MR for the detection of the acute (6 hour) and sub-acute (1 week) myocardial changing after coronary artery microembolization (CME) in swine and to assess the value of 3.0-T cardiac MR imaging in the detection of CME.Materials and Methods:Sixteen swines, of either sex (body weight 25-30 kg), were devided into two groups randomly:CME group and control group. CME was induced in CME group by injection of 120,000 microspheres (42μm in diameter) into the left anterior descending (LAD) coronary artery through micro-catheter and 10ml saline was injected into the LAD in control group. Two swines died during surgical preparation. Cardiac MR imaging was performed by using 3.0-T system (Magnetom Verio, Siemens AG, Germany) before and repeated six hours and one week after CME. The MR sequences included:TrueFISP cine sequence combining with parallel imaging technique TGRAPPA (Temporal Generalized Auto-calibrating Partially Parallel Acquisitions)(accelerated factor 3) for the measurement of cardiac function, TSE-T2WI with scanning mode BLADE for the detection of myocardial edema and TurboFLASH sequence for the discovery of myocardial microinfarction. After all the cardiac MR scanning was completed, the animals were then sacrificed for myocardial pathologic analyses (NBT and HE).Results:No abnormality was found on coronary angiography (DSA) before and after CME. In control group (2 swines), the left ventricular wall motion had no significant changes and no delayed-enhancement segments were found at 6th hour and 1 week. In CME group, systolic function of anterior wall and anterior-septal was impaired in all 12 swines variably at 6th hour and recovered slightly one week later. The left ventricular ejection fraction (LVEF) measured by cine sequence decreased significantly from (64.77±5.75)% at baseline to (54.15±3.77)% at the 6th hour (P <0.05), and improved to (55.65±8.41)% one week later (baseline vs 1 week, P<0.05). The left ventricular end diastolic volume (LVEDV) enlarged significantly from 39.27±7.35ml at baseline to 48.22±5.13ml at sub-acute stage after CME (baseline vs 1 week, P<0.05). The left ventricular end systolic volume (LVESV) was 13.92±3.63ml, 20.02±4.20ml,21.50±5.33ml at the three time points respectively and significant difference were found in both baseline vs 6th hour and baseline vs 1 week. In T2WI-TSE sequence, the myocardial hypersignal was only observed at 1 week in 2 swines in areas adjacent to apex. Myocardial delayed enhancements were found on LV short-axis section in all swines during acute stage. However, the delayed enhanced areas disappeared in 10 swines and only existed in 2 swines, which were consistent with NBT pathological findings. Coronary microembolizations were demonstrated pathologically.Conclusion:Coronary microembolization caused significant decline in LVEF at 6th hour and improved slightly at 1 week. The LVEDV and LVESV were persistent enlarged. Myocardial delayed enhancement were found in all swines at 6th hour and disappeared or diminished one week later after CME, which were consistent with NBT pathological findings.3.0T cardiac MR imaging has potential to help reliably quantify the changes of cardiac function and to detect the microinfaction at acute and sub-acute stage. However, improved scanning technique is needed to detect the myocardial edema after CME.