【作者】 于静；

【导师】 于铁链；

【作者基本信息】 天津医科大学 ， 影像医学与核医学， 2014， 博士

【摘要】 第一部分心脏能谱CT:CT值,碘定量和能谱曲线评估猪模型急性心肌梗死背景及目的：-近年来问世的能谱CT,作为一种新工具,对于研究和评估心血管疾病,具有传统混合能量CT难以比拟的应用潜力。本研究的第一部分目的是应用心脏能谱CT定量评估猪急性缺血-再灌注型MI的能力并确定显示MI的最佳keV和延时时间。对象和方法：12只实验动物猪通过介入球囊扩张造成前降支第一对角支以远梗阻,行90分钟阻塞后撤出球囊进行心肌再灌注。猪急性缺血-再灌注型MI模型成功建立4±1天后,行心脏能谱CT冠状动脉造影(computed tomography coronary angiography, CTCA)。以5keV为间距,选取从40keV至140keV的21个单能图像分别测量MI图像的信噪比(signal to noise ratio, SNR)、对比噪声比(contrast to noise ratio, CNR)和噪声,寻找观察梗死区的最佳单能量。之后阐述最佳单能图像、高电压图像(高kVp图像)及碘基图像上观察MI时各自的优势与不足。通过单因素方差分析(analysis of variance, ANOVA)或T检验,观察心肌梗死区、危险区及正常心肌的密度(CT值)、碘浓度或能谱曲线在最佳单能图像、高电压图像和碘基图像上是否具有差别。能谱CTCA图像采集完毕后,对比剂注入1,3,5,10,和15min时,再分别采集心脏能谱CT延迟增强图像,观察梗死区在最佳单能图像上的变化,选择评估MI的最佳延时时间。能谱CT检查结束后,行心脏核磁检查,通过T2高信号区,确定本研究中的危险区。影像学检查结束后,通过注射过量肌松药处死实验动物,切取4mm的心脏短轴位标本进行TTC染色,证实心肌MI的部位。结果：在40keV-140keV的范围中,70keV-75keV图像是观察心肌梗死的最佳单能图像,其SNR和CNR均较高,而噪声较低。行能谱CTCA检查时,梗死区、危险区及正常心肌在最佳单能图像(70keV)、高电压图像和碘基图像的CT值或碘浓度均存在显著差异◇<0.05),其中70keV图像的SNR、CNR较高,而噪声最低。心肌梗死区、危险区及远处正常心肌的能谱曲线,其起始点(即40keV的CT值)及经过对数转换的斜率均存在显著差异。在70keV图像上,梗死区在对比剂注入后5-10min强化最为明显,且与周围组织的对比差异较大。结论：心脏能谱CTCA可通过CT值测量评估急性MI,70keV图像更具有优势,因其信噪比和对比噪声比较高。此外,碘浓度和能谱曲线也可以辅助评估MI。对比剂注入后5～10分钟被认为是能谱CT延时增强检查观察猪急性MI的最佳延时时间。第二部分急性缺血-再灌注型心肌梗死：能谱CTCA、心肌灌注SPECT和TTC染色背景及目的：通过分析能谱CTCA与TTC染色定性MI心肌段的一致性,及其与心肌灌注SPECT定量评估心肌灌注的相关性及一致性,来探讨能谱CTCA评价MI及心肌灌注的能力。对象和方法：12只实验动物猪(体重20.23±1.23kg,5只雄性,月龄6.0±0.7月)进行急性MI模型制作,通过球囊扩张的介入方法,堵闭前降支第一对角支以远,90分钟后撤出球囊进行心肌再灌注。猪急性缺血-再灌注型MI模型成功建立4±1后,行心脏能谱CTCA、心肌灌注SPECT检查。能谱CTCA图像、TTC染色大体病理标本均按照美国心脏学会标准心肌分段确定各自诊断标准的梗死心肌段,通过Cohenκ一致性检验分析能谱CTCA和TTC染色确定MI的一致性,此外采用受试者工作特征(Receiver operating characteristic, ROC)曲线评估心脏能谱CTCA定义TTC证实的MI节段的能力。能谱CTCA和心肌灌注SPECT通过测量17段心肌灌注减低率,进行Pearson相关性和组内相关系数(intraclass correlation coefficient, ICC)一致性分析,探讨能谱CTCA评估急性MI的心肌灌注的能力。结果：12只MI建模成功的实验动物中,2只在造模结束后死于室颤,1只在能谱CTCA检查中发生室颤死亡,其余9只(年龄5.3±0.6月;体重,20.17±1.35kg；3只雄性)均完成了能谱CTCA、心肌灌注SPECT检查和TTC病理染色。能谱CTCA及TTC染色分别识别各自诊断标准的梗死心肌段53段和58段,Cohen κ分析显示能谱CTCA和TTC染色识别的梗死心肌段一致性较好(κ=0.681,p<0.001),且ROC曲线分析显示能谱CTCA识别心肌梗死节段的效能很高(敏感性,0.810;特异性,0.937；阳性预测值,0.887；阴性预测值,0.880,准确性,0.874；p<0.001)。能谱CTCA检查共识别53个灌注减低率大于30%的心肌段,其中TTC染色证实的梗死心肌段47段,心肌平均灌注减低率为21.27%；心肌灌注SPECT共识别51个灌注减低率大于30%的心肌段,TTC染色证实的梗死心肌段为44,心肌平均灌注减低率为19.82%,与能谱CT灌注减低率无显著统计学差异(p=0.06)。能谱CTCA和心肌灌注SPECT的心肌灌注结果相关性较高(r2=0.837,p<0.001),一致性较好。结论：能谱CTCA可以识别梗死心肌段,其诊断效能较高,识别的梗死心肌段与TTC病理染色确定的梗死心肌段一致性较好,从而可对MI进行定性诊断,但会遗漏小病灶MI。此外,能谱CTCA在70keV图像上测得的心肌灌注减低率与心肌灌注SPECT测量结果具有显著相关性,一致性很好。第三部分碘定量与TUNEL染色：无-复流区,梗死区和正常心肌背景及目的：能谱CT有望通过基物质图像的碘定量无创性评估心肌灌注情况。本研究的目的是通过能谱CT评估急性心肌缺血-再灌注型MI猪模型中心肌无-复流区、梗死区和正常心肌的碘浓度差异,并探讨其与免疫荧光病理染色(TUNEL)染色确定的心肌细胞凋亡、坏死的相关性。对象和方法：根据医院动物饲养和使用委员会要求,中华小型猪接受了人性关怀。12只实验动物猪(5只雄性,6.0±0.7月龄,体重20.23±1.23kg)进行经皮冠状动脉介入(Percutaneous Coronary Intervention, PCI)干预,通过球囊扩张制作急性缺血-再灌注型MI模型。急性缺血-再灌注型MI模型成功建立44-1天后,实验动物猪行能谱CTCA检查和5分钟后的延时增强检查来评估MI。观察延时增强图像,选取出现无-复流现象的实验对象。测量无-复流区、梗死区及正常心肌的碘浓度。影像学检查结束后,静脉注射过量维库溴铵注射液处死动物,取出心脏并切成4mm厚的短轴位切片。结合能谱CT延时增强图像及大体标本的结果,选取无-复流区、梗死区及正常心肌连续切取组织制作切片进行免疫荧光染色。细胞凋亡通过aterminal deoxynucleotidyl transferase--mediated dUTP Nick-End Labeling (TUNEL)染色进行标记,TUNEL染色按照厂家使用说明通过In Situ Cell Death Detection Kit Fluorescein进行染色。采用Leica TCS SP5共聚焦显微镜,随机选择3个视野,采集免疫荧光染色图像。TUNEL染色阳性细胞核通过图像分析软件‘’Image-Pro Plus Version6.0"进行计数。结果以阳性细胞核计数／图像面积表示。组间差异通过ANOVA最小显著差异(least significant difference, LSD)检验进行分析。无-复流区、梗死区及正常心肌三个区域的碘浓度和细胞凋亡进行Pearson相关分析。结果：无-复流区、梗死区及正常心肌的碘浓度分别为1043±282ug/cm3,1867±344ug/cm3,和3507±331ug/cm3,单因素ANOVA分析显示三者之间存在显著差异(p<0.001)。通过免疫荧光TUNEL染色观察到,无-复流区、梗死区及正常心肌的心肌细胞凋亡存在统计学差异,分别为(2661±231)/mm2,(2270±241)/mm2and (27±22)/mm2.心肌的碘浓度与细胞凋亡呈负相关(r2=0.871,p<0.001)。结论：心脏能谱CTCA可通过碘定量区别无-复流区与心肌梗死区、正常心肌。此外,心脏能谱CTCA测量急性MI猪模型的心肌碘浓度与免疫荧光TUNEL染色所示的细胞凋亡呈显著负相关。将来,心脏能谱CTCA的碘定量有可能为危险分级提供有价值的信息。更多还原

【Abstract】 Part Ⅰ Cardiac Spectral CT:attenuation density, iodine quantification and spectral curve of acute myocardial infarction in swinesBackground and Objective:As a new experiment introduced in recent year, spectral CT had much potential use compared with conventional CT in the research and evaluation of cardiovascular disease. The current study was to quantitatively assess acute ischemic-reperfused MI in swine model by cardiac spectral CT. Methods:12swines underwent left anterior descending coronary artery (LAD) occlusion just distal to the first diagonal branch by balloon angioplasty.90minutes later, the balloon was withdrawn to undergo myocardial reperfusion. Cardiac spectral CT coronary angiogram (CTCA) was performed after4±1days of successful acute MI model establishment. The signal to noise (SNR), contrast to noise ratio (CNR) and noise were compared among the21monochromatic energy images from40keV to140keV at an interval of5keV. The optimal energy image for MI delineation was chosen base on it. The characteristics of MI on optimal keV, high voltage and iodine density images were evaluated. The differences of CT value, iodine concentration (IC), and spectral curve among MI, risk area and remote myocardium were investigated by analysis of variance (ANOVA) or t test.1,3,5,10, and15minutes after contrast material administration, the delayed enhancement images were acquired to observe the evolution of MI on optimal keV images of spectral CT. The best delay time for MI evaluation was determined consequently. Addtionally, cardiac magnetic resonance imaging was performed to help determine the area at risk (high signal on T2images) in the current study. After radiologic examination, overdose muscle relaxant was given to swine. The heart was excised and sliced in4mm short axis. Each slice was stained by TTC to delineate the location and extent of MI. Results:From40keV-140keV, the70keV-75keV images were defined the optimal keV images for MI evaluation due to higher SNR, CNR, and lower noise. Significant differences of CT value, IC on spectral CTCA were found among infarction, risk area and remote myocardium on optimal keV (70keV), high voltage and iodine density images (p<0.05), especially on70keV images with higher SNR, CNR and lower noise (p<0.01). The start values and the log transformed slopes of infarct myocardium, risk area and remote myocardium had significant differences. The infarction enhancement was best shown in images obtain5～10minutes after contrast material administration on70keV images, the contrast between infarction and adjacent myocardium was more prominent. Conclusion:The spectral CTCA could assess acute myocardial infarction by CT value measurement.70keV images with higher SNR, CNR and lower noise was preferable. In addition, the IC and spectral curve also help the infarction evaluation.5～10minutes after contrast material administration was regard as the best delay time to demonstrate MI in swine. Part II Acute ischemic-reperfused MI:the comparison between spectral CTCA and myocardial perfusion SPECT as well as TTC stainBackground and Objective:The current study was to investigate the ability of spectral CTCA to evaluate MI and myocardial perfusion in swine model by comparing the cardiac spectral CTCA with TTC stain qualitatively and with myocardial perfusion SPECT quantitatively in MI recognition. Methods:12acute MI model of swines (weight,20.23±1.23kg,5male,6.0±0.7months) were established. LAD just distal to the first diagonal branch was occluded by balloon angioplasty.90minutes later, the balloon was drawn back to undergo myocardial reperfusion. Spectral CTCA, myocardial perfusion SPECT were performed after4±1days of successful acute MI model establishment. The images of spectral CTCA and TTC stain identify MI segments by their own criteria. Myocardial segmentation was according to American Heart Association standard. Cohen κ test was used to analysis the consistency between spectral CTCA and TTC stain in MI segment differentiation. In addition, receiver operating characteristic (ROC) curve was used to investigate the ability of spectral CTCA to differentiate MI segments confirmed by TTC stain. The myocardial defect of17segments measured by spectral CTCA and myocardial perfusion SPECT underwent Pearson correlation and intraclass correlation coefficient (ICC) consistency analysis. Accordingly, the ability of spectral CTCA to evaluate myocardial perfusion had been investigated. Results:12swines underwent acute MI model establishment successfully.2of them died of ventricular fibrillation after PCI.1of them died of ventricular fibrillation during the spectral CTCA examination. The rest9swines (5.3±0.6months,20.17±1.35kg,3male) all underwent spectral CTCA and myocardial perfusion SPECT examination.53and58myocardial segments were identified as MI by spectral CTCA and TTC stain respectively according to their diagnostic criteria. Cohen κ test demonstrate moderate consistency between spectral CTCA and myocardial perfusion SPECT in MI segment differentiation (κ=0.681, p<0.001). ROC curve showed high diagnostic accuracy of spectral CTCA to differentiate myocardial infarct segments (sensitivity,0.810; specificity,0.937; positive predictive value,0.887; negative predictive value,0.880and accuracy,0.874, p<0.001). Spectral CTCA identified53segments which myocardial perfusion defect larger than30%.47of the53segments were confirmed as MI segment by TTC stain. The mean perfusion defect of153segments in9swines was21.27%. The myocardial perfusion SPECT identified51segments which myocardial perfusion defect larger than30%.44of the51segments were confirmed as MI segment by TTC stain. The mean perfusion defect of153segments in9swines was19.82%, with no significant statistically difference with spectral CTCA (p=0.06). The correlation between spectral CTCA and myocardial perfusion SPECT was high in myocardial perfusion assessment. Conclusion:The spectral CTCA could identify MI segment confirmed by TTC stain with high diagnostic ability. In other words, spectral CTCA could assess acute MI qualitatively; however, the small MI may be missed by spectral CTCA examination. In addition, significant correlation and good consistency were investigated between70keV images of spectral CTCA and myocardial perfusion SPECT in myocardial perfusion defect measurement. Part Ⅲ Evaluation of myocardial infarction by iodine quantification and TUNEL stainBackground and Objective:The spectral CT could be used to evaluate myocardial perfusion by iodine quantification on the basic material decomposition image noninvasively. The purpose of this study was to evaluate IC of the no-reflow region, infarction and remote myocardium of acute ischemic-reperfused MI model in swines by spectral CT. Its correlation between iodine quantification and tunnel stain was analyzed. Methods:In compliance with the institutional animal care and use committee, Chinese mini swines received human care.12swines (5male,6.0±0.7months old, weight,20.23±1.23kg) underwent Percutaneous Coronary Intervention (PCI) to produce acute reperfusion MI by balloon dilation.4±1days after acute ischemic-reperfused MI model establishment, cardiac spectral CTCA and late enhancement imaging at5minutes after contrast material injection were performed to evaluate MI. The objects with no-reflow phenomenon on late enhancement imaging were chosen. The IC in no-reflow region, infarction and remote myocardium was measured. After radiologic examination, overdose vecuronium bromide was given in venous, the heart were excised and sliced into4mm short axis slices. With the consideration of spectral CT images and gross specimen, serial cutting sections in the no-reflow region, infarction and remote myocardium were used for immunofluorescent staining. Cell apoptosis was detected by a terminal deoxynucleotidyl transferase-mediated dUTP Nick-End Labeling (TUNEL) stain by using the In Situ Cell Death Detection Kit Fluorescein according to the manufacturer’s instructions. A Leica TCS SP5confocal microscope were used for acquisition of immunofluorescent staining images in3random field of view. The TUNEL positive nuclei were counted on each section by Image analysis system "Image-Pro Plus Version6.0". The results were represented as positive nuclei count/area and the mean value were recorded. The differences among groups were compared by using ANOVA and least significant difference test. The correlation between IC and cell apoptosis of no-reflow region, infarction and remote myocardium was test by Pearson correlation analysis. P<0.05was considered statistically significant. Results:The iodine concentrations were1043±282ug/cm3,1867±344ug/cm3, and3507±331ug/cm3for of no-reflow region, infarction and remote myocardium respectively. There was significant among them (p<0.001). The apoptosis cells in no-reflow region, infarction and remote myocardium were (2661±231)/mm2,(2270±241)/mm2and (27±22)/mm2respectively. Significant differences were detect among them (p<0.001). There was significant inverse correlation between iodine concentration and cell apoptosis (r2=0.871, p<0.001). Conclusions:The IC calculated on spectral CT images could differentiate no-reflow region from infarction or remote myocardium. Furthermore, the myocardial IC on spectral correlated with the TUNEL stain of apoptosis cell in acute MI of swine models. The iodine quantification on cardiac spectral CT may add valuable information for risk stratification in the future.更多还原