Myocardial Ischemia And Viability: Assessment With Combination Of Multiple Cardiac Magnetic Resonance Imaging Techniques

Part one: The MR Imaging techniques of Cardiac Cine and Delay-Enhancement Imaging, the Analysis of the Artifact in Myocardial perfusion scan1. Assessment of ventricular function and image quality of single breath-hold and free-breath real-time TrueFISP with TSENSE eine MR imaging.Objective: To assess the impact of parallel imaging algorithms on image quality and volumetric accuracy of cine magnetic resonance imaging (MRI) with time-adaptive sensitivity encoding (TSENSE) algorithms in a single breath-hold or free breath.Material and Methods: Fifteen patients underwent steady-state free precession (SSFP) cine MRI for short axis images of global heart scans. All sequences include true fast imaging with steady-state precession (TrueFISP) combined with GRAPPA, TrueFISP with TSENSE of various acceleration factors ranged from R=2 to 4, and free-breath real-time TSENSE. Volumetric assessment including ejection fraction (EF), end-diastolic volume (EDV) and end-systolic volume (ESV) was performed in each sequence. Contrast-to-noise ratio (CNR) binding subjective visual evaluation was used to assess the imaging quality. Results: No significant difference among these sequences in measurement of EF,EDV and ESV was observed, compared with TrueFISP sequence with GRAPPA. All MR images could be diagnosed, among which the quality of GRAPPA and TSENSE (R=2) was better. Conclusion: A single breath-hold or free-breath SSFP cine MRI with TSENSE can precisely evaluate cardiac volumetry. SSFP cine MRI with TSENSE (R=3) will be fit for clinical routine work.2. Comparison of three-dimensional (3D) and two-dimensional (2D) delayed enhancement MR imaging.Objective: To prospectively evaluate the correlation between a single breath-hold 3D and a traditional several breath-hold 2D delayed enhancement MR imaging sequence in assessment of myocardial infarction size and the quality of the images. Methods and Materials: 15 of myocardial infarction patients were underwent MR scans including a single breath-hold 3D inversion-recovery fast low-angle shot (FLASH) sequence and several breath-hold 2D turbo-FLASH as the reference standard. Paired samples statistics was used to compare the ratio of the infarction areas of two sequences. Two way ANOVA was used to assess the Contrast-to-noise ratio (CNR) of 3D,2D-magnitude and 2D-phase-sensitive images. Meanwhile subjective visual evaluation was also performed to compare the quality of the images. Results: The infarction ratio determined by using 3D MR imaging (31.28%±19.70%) has no significant difference with that of 2D MR imaging ( 30.91%±19.14%). The CNR was significantly higher with 3D and 2D-magnitude images than with 2D-phase-sensitive images. However, 2D-phase-sensitive images demonstrated the greatest contrast between normal myocardium and the infarction in subjective visual evaluation. On the other hand, the background noise of 2D-phase-sensitive images was the highest.Conclusion: The ratio of myocardial infarction areas obtained with the 3D MR imaging sequence is accurate, and the quality of images is good. As reducing the acquisition time, the sequence is fit for clinical application. But the imaging sequence is easily affected by the choice of TI and the state of breath-hold.3. The endocardial "dark rim" artifact in MRI myocardial perfusion of healthy pigs.Objective: To analyze the causes and the appearance of the endocardial "dark rim" artifact in MR first-pass myocardial perfusion imaging. Materials and Methods: First-pass contrast material-enhanced MR imaging was performed at rest and after the infusion of dipyridamole in 11 healthy pigs that were proved to be normal in the coronary angiography before MR scans. The association between the band artifact and the influence factors including the susceptibility effects from the gadolinium bolus,cardiac motion,the direction of the phase-encoding(in-plane spatial resolution) and slice thickness (through-plane spatial resolution) was analyzed. Binary Logistic Regress statistical analysis was used to evaluate the relationship of them. Results: The dark rim artifact was due to the combination of susceptibility effects,cardiac motion and in-plane spatial resolution. Among them, the influence of susceptibility effects was the largest, whose OR (odds ratio) was 5.735. The "dark rim" artifact often arised in tandem with the arrival of the bolus of gadolinium in the left ventricle. It typically preceded myocardial tissue uptake and was transient in nature. However, the effect of slice thickness was limited. Conclusion: The endocardial "dark band or rim" artifact was due to many factors, while the susceptibility effect played the dominating role. The recognization of the appearance of the artifact will be helpful for differentiating with the true perfusion defects.Part Two: Assessment of Myocardial Ischemia and Viability: the Combination of Multiple Cardiac MR Imaging Techniques in Chronic Myocardial Ischemic Model of PigsObjective: To evaluate the myocardial ischemia and viability using the combination of multiple cardiac MR imaging techniques including cine,first-pass myocardial perfusion imaging, delayed-enhancement(DE) MR imaging and low-dose dobutamine test in the chronic myocardial ischemia model of pigs. These imaging techniques are to form various compounding. The diagnostic accuracy is to be assessed for each protocol respectively. Methods and Material: Chronic myocardial ischemic animal model in 20 pigs were established with setting the Ameroid contract ring in left circumflex coronary. MR scans were performed after 4～6 weeks in models. The scan procedure included rest cine,rest and dipyridamole stress myocardial perfusion imaging, delayed-enhancement imaging and low-dose dobutamine test. Coronary angiography was performed in alive models before putting pigs to death. Electronic microscope scan, triphenyl tetrazolium chloride (TTC), haematoxylin & eosin (HE) and poley staining were processed after pigs' death. Seven of the models underwent the rest ^99mTc-MIBI myocardial perfusion combined with ¹⁸F fluorodeoxyglucose (FDG) metabolic nuclear imaging. Using histopathologic results as the reference standard of myocardial ischemia, the detection of the ischemia with various protocols involving rest cine alone,rest or/and stress myocardial perfusion imaging,DE-MR and rest cine binding DE-MR was evaluated with ROC curve. The signal intensity-time curve was used to semiquantitative analysis of the perfusion. Using nuclear imaging linking to histopathologic results as the reference standard of myocardial viability, the diagnostic accuracy of different protocols was judged, which included low-dose dobutamine test alone,rest cine+DE-MR, rest cine+DE-MR+stress perfusion imaging and DE-MR+low-dose dobutamine test. Results: The modeling was successful in 13 pigs, which were proved to be significant stenosis of LCX by coronary angiography. In analysis of myocardial ischemia (78/208), rest cine combined with stress first-pass perfusion MR imaging yielded the highest sensitivity of 90.3%, specificity of 88.7%, and accuracy of 92.7% among all methods. The upslope of the signal intensity-time curve was most valuable in differentiating ischemic myocardium from normal myocardium. In assessment of myocardial viability, 112 segments were classified as the pure infarction (n=6), infarction intermixed viable myocardium (n=14), absolute viable myocardium (n=23) and normal myocardium (n=69). The combination of rest cine,stress first-pass perfusion imaging and DE-MR greatest highly correlated with the results of nuclear imaging binding histopathology (Kappa=0.759). If the segments were classified as 3 types (the infarction,viable myocardium and normal myocardium), the Kappa value of this combination would increase to 0.895. The Kappa values of other protocols were all under 0.75. Conclusion: In assessment of the myocardial ischemia and viability, combination of multiple cardiac MR imaging techniques will improve the diagnostic accuracy. Rest cine combined with DE-MR and stress first-pass perfusion imaging is the best choice.Part Three: The Clinical Application of Multiple Cardiac MR Imaging Techniques in Diagnosis of CADObjective: To evaluate the clinical application value of the current multiple cardiac MR imaging techniques. Methods and Material: 42 patients with coronary artery disease (CAD) underwent MR scans. The MR imaging protocol included rest cine MR imaging,rest first-pass myocardial perfusion and delayed-enhancement(DE) MR imaging. Myocardial segments were divided into four groups according to the degree of stenosis in the supplying artery. The parameters of perfusion signal intensity-time curve, including upslope,peak signal intensity, and time to peak signal intensity were analyzed for each segment in different groups. Qualitative analysis of first-pass perfusion imaging was also used to evaluate myocardial ischemia. The infarction ratio obtained with DE-MR in 25 cases of acute myocardial infarction. The correlation between the infarction ratio with peak value of creatine kinase (CK) and cTnT was analyzed using a generalized linear model. According to the recovery of wall systolic function as the reference standard of myocardial viability, the transmural extend of myocardial infarction demonstrated by DE-MR and rest cine MR imaging were used to differentiate the component of the myocardium. Results: The 4 patients were excluded from signal intensity-time curve analysis because of poor-quality MR images. Signal intensity upslope in patients with normal coronary arteries at angiography was significantly higher than that in patients with significant stenosis of coronary arteries. The diagnostic value of signal intensity upslope was the greatest among these parameters. With quality analysis of first-pass perfusion imaging, ischemic myocardium (the degree of stenosis＞50%) was detected with a sensitivity of 83.1% and a specificity of 45.5%. There was a significant correlation between ratio of infarction areas and peak CK values (r=0.882, P＜0.05), as well as the ratio and peak cTnT values (r=0.677, P＜0.005). 10 patients with AMI were followed. The transmural extend of infarction＜25%,＜50% and＜75% to detect the viable myocardium yielded the sensitivity of 72.2%, 78.2%, 88.7% and the specificity of 66.7%,63%,55.6% respectively. Combining rest cine MR imaging to evaluate the wall systolic function, the sensitivity was up to 92.5%, but the specificity did not improve. Meanwhile, six segments without delayed enhancement and abnormal wall motion before PCI, were found to be hypokinetic postoperate. Four of them were depicted with low signal intensity in first-pass perfusion MR imaging before PCI.Conclusion: Cardiac MR imaging techniques at rest can provide some information for the diagnosis of CAD. The first-pass perfusion imaging using semiquantitative analysis can help to detect the ischemic myocardium. And it can evaluate microvascular obstruction for acute myocardial infarction. DE-MR can demonstrate the location and size of the infarction precisely, and the transmural extend of the delayed-enhancement binding rest wall motion assessment has high sensitivity in the detection of viable myocardial segments.