【作者】 孙旭东；

【导师】 黄子扬； 陈良龙；

【作者基本信息】 福建医科大学 ， 内科学， 2011， 博士

【摘要】 目的:双源CT、3.0T磁共振对比增强全心冠状动脉成像(增强全心冠脉成像)及经胸多普勒超声心动图冠状动脉血流成像均为新近发展的无创性冠状动脉检测新技术,但迄今尚未见同步对比此三者诊断冠状动脉明显狭窄能力的前瞻性研究。本文探讨双源CT、3.0T磁共振增强全心冠脉成像及经胸多普勒超声心动图对冠状动脉明显狭窄的诊断价值。方法: 51例疑诊或确诊冠心病拟行冠状动脉造影检查的连续受检者在行冠状动脉造影检查前一天内行同步双源CT、3.0T磁共振增强全心冠脉成像及经胸多普勒超声心动图冠状动脉检查,以冠状动脉造影为金标准进行诊断试验,采用盲法评价三者诊断冠状动脉明显(≥50%)狭窄的准确性。结果:1.双源CT、3.0 T磁共振和超声心动图分别有486段(486/512,96.0%)、479段(479/512,93.6%)及398段(398/512,77.7%)可供评价,双源CT及3.0 T磁共振可评价节段相似,超声心动图可评价节段少于双源CT及3.0 T磁共振。2.以冠状动脉造影结果为对照,以冠状动脉血管为单位的分析结果是:3.0T磁共振对比增强全心冠脉成像与双源CT对诊断冠状动脉明显狭窄诊断准确性相似,虽有稍逊于双源CT的趋势,但差异并无统计学意义,P均>0.05;多普勒超声心动图对冠状动脉明显狭窄诊断的敏感性、特异性、阳性预测值及阴性预测值比双源CT及3.0T磁共振对比增强全心冠状动脉成像都低,差异有统计学意义,P均<0.05,其中双源CT、3.0T磁共振对比增强全心冠状动脉成像和多普勒超声心动图诊断冠脉明显狭窄的敏感性分别为94%、88%及65% ;特异性分别为90%、88%、72%;阳性预测值为82%、78%、52%;阴性预测值分别为97%、95%、81%。3.以冠状动脉节段为单位的分析结果是:3.0T磁共振对比增强全心冠脉成像与双源CT对诊断左主干+左冠状动脉、左旋支及右冠状动脉明显狭窄诊断准确性相似,虽有稍逊于双源CT的趋势,但差异并无统计学意义,P均>0.05;多普勒超声心动图对诊断左主干+左冠状动脉、左旋支及右冠状动脉明显狭窄的敏感性、特异性、阳性预测值及阴性预测值比双源CT、3.0T磁共振对比增强全心冠状动脉成像都低,差异有统计学意义,P均<0.05,其中双源CT、3.0T磁共振对比增强全心冠状动脉成像和多普勒超声心动图对左主干+左前降支明显狭窄诊断的敏感性分别为95%、89%、79% ;特异性分别为90%、90%、78%;阳性预测值为86%、85%、68%,阴性预测值分别为97%、93%、86%;对左回旋支诊断的敏感性分别为95%、85%、46% ,特异性分别为95%、92%、68%,阳性预测值为90%、79%、33%,阴性预测值分别为97%、95%、79%;对右冠状动脉诊断的敏感性分别为94%、88%、65% ;特异性分别为85%、82%、69%;阳性预测值为70%、71%、52%;阴性预测值分别为97%、93%、79%。经胸超声心动图对左主干+左前降支明显狭窄诊断敏感性最高(79%)、右冠状动脉其次(65%),左旋支最低(46%)。4.经胸超声心动图检测左前降支明显狭窄处舒张期峰值血流速度为3.6±1.9M/S,狭窄处与狭窄前峰值血流速度比率为3.3±1.2,左旋支明显狭窄处舒张期峰值血流速度为3.0±0.5M/S,狭窄处与狭窄前峰值血流速度比率为2.8±0.6,右冠状动脉明显狭窄处舒张期峰值血流速度为2.8±0.6M/S,狭窄处与狭窄前峰值血流速度比率为2.7±0.3,左前降支、左回旋支与右冠状动脉明显狭窄处舒张期峰值血流速度以及狭窄处与狭窄前峰值血流速度比率差异无统计学意义。结论:双源CT、3.0T磁共振增强全心冠脉成像对冠状动脉明显狭窄均有相似的高度的诊断准确性,虽然双源CT的各项评价指标均略优于3.0T磁共振,但差异均无统计学意义。超声心动图由于较低的成功显示率以及较少的可评价节段,诊断准确性较低。3.0T磁共振增强全心冠脉成像的优势在于无Ⅹ线及碘对比剂的潜在危害,多普勒超声心动图经济便携且能提供丰富的冠状动脉血流动力学信息,可作为对双源CT与3.0T磁共振增强全心冠脉成像的有益补充。目的:识别心梗的透壁程度(The transmural extent of infarction ,TME)兼具指导临床再灌注治疗及评判预后的双重意义。目前,无创性的超声心动图二维斑点追踪成像(2D-speckle tracking imaging,2D-STI)技术是否可用于识别陈旧性心梗透壁程度、具体方法及其实际应用价值均尚未明确。本研究拟探讨2D-STI技术识别陈旧性心梗患者左室整体心梗大小与节段水平透壁程度的可行性及其有效方法。方法:对43位临床确诊的陈旧性心肌梗死的连续性患者进行2D-STI、磁共振延迟强化成像(delayed-enhanced magnetic resonance imaging,DE-MRI)及生化检查的随机盲法同步临床试验。2D-STI左室整体水平指标包括整体纵向应变(Global longitudinal strain,GLS)、旋转及扭转指标;左室节段水平指标包括各节段舒张期室壁厚度及各节段应变指标,后者包括短轴上各节段径向应变(radial strain,SR),圆周应变(circumferencial strain,SC),旋转角(rot);以及长轴上各节段纵向应变(longitudinal strain,SL)。整体水平心肌梗死大小包括心肌梗死容积率及梗死质量;节段水平透壁心肌梗死指DE-MRI延迟强化区≥50%节段总容积。<50%为非透壁心肌梗死。左室心尖各节段依TME分成透壁组及非透壁组,生化检查指标包括血液超敏C反应蛋白、NT Pro BNP及同型半胱氨酸测定。结果:①在左室整体水平, GLS与左室整体心梗容积率显著相关(P=0.008),相关系数等于0.620;GLS和心尖最大旋转率是左室整体心梗容积率的显著预测因子(P分别等于0.005及0.014),标准回归系数(Beta)分别为0.720及0.592;GLS是左室整体心梗质量的预测因子(P=0.024),标准回归系数(Beta)为0.545,各生化指标不是左室整体心梗容积率及质量的预测因子;②在左室节段水平:与左室心尖段心肌梗死非透壁组相比,左室前壁心尖段透壁组舒张末期室壁厚度、SR(ES),SC(ES),SR Peak G,SL(ES),SC Peak,SL Peak G,SL Peak及SC Peak G(依ROC曲线下面积降序排列,下同)差异有统计学意义(P均<0.05);左室侧壁心尖段SC(ES),SC Peak G,SC Peak及SL(ES)差异有统计学意义(P均<0.05);前间隔心尖段SL Peak ,SL Peak G, SC Peak ,SL(ES),SC(ES)及SR(ES)差异有统计学意义(P均<0.05);左室后壁心尖段SR Peak G,SR(ES)及SL(ES)差异有统计学意义(P均<0.01);左室下壁心尖段SC Peak G,SC Peak及SC(ES)差异有统计学意义(P均<0.01);后间隔只有舒张末期室壁厚度差异有统计学意义(P=0.007) ;而各节段旋转角指标差异均无统计学意义,③ROC曲线分析显示舒张末期室壁厚度为判断左室前壁及后间隔心尖段透壁心梗的最佳指标,前者临界值5.5MM时诊断透壁心肌梗死的敏感度为83.3%、特异度为85.7%,后者临界值5.5MM时敏感度为100%、特异度为61.5%;SC(ES)为左室侧壁心尖段最佳指标,临界值-3.0800时,敏感度为100%、特异度为94.4%;SL Peak S为左室前间隔心尖段最佳指标,临界值-5.4650时敏感度为71.4%、特异度为100%;SR Peak G为左室后壁心尖段最佳指标,临界值12.265时敏感度为100%、特异度为80.0%;SC Peak G及SC Peak S均为左室下壁心尖段最佳指标,临界值-5.3200时,敏感度为100%、特异度为88.8%。结论:①在左室整体水平, 2D-STI技术的左室整体纵向应变可准确评价陈旧性心肌梗死整体心梗大小;左室心尖旋转率是较左室整体扭转角或扭转率更好的评价指标。②在左室节段水平,发现纵向应变、径向应变、圆周应变以及/或室壁厚度指标分别可检出左室心尖不同节段的透壁心梗,但不同应变指标的识别能力表现各异,这可能是左室各节段心肌方向、透壁应变不同的复杂性的真实反映。目的:术前准确识别冠状动脉解剖变异类型是大动脉转位(transposition of great arteries, TGA)根治术即动脉调转术(arterial switch operation, ASO)成功的关键因素之一。本文探讨术前经胸超声心动图与双源CT检查对识别冠状动脉解剖变异类型的诊断价值及提高超声心动图诊断准确性的方法。方法:将164例准备行ASO患儿术前超声心动图检查与术中所见进行对比,其中大动脉转位室间隔完整(TGA with intact ventricular septum ,TGA/IVS)患者49例,大动脉转位伴室间隔缺损(TGA with ventricular septum defect ,TGA/VSD) 77例,右室双出口伴TGA(Double outlet right vetricle with TGA,DORV/TGA) 38例。164例中的53例连续TGA受检者在行ASO手术前三天内同时行双源CT检查。全部病例采用盲法以手术所见为金标准进行诊断试验。结果: 164例患者中,冠状动脉发育正常124例(75.61%),冠状动脉畸形40例(24.39%),其中TGA/VSD及DORV/TGA中冠状动脉畸形32例(80.00%),TGA/IVS冠状动脉畸形8例(20.00%),两者差异有统计学意义(P<0.05)。共发现10种异常类型,其中冠状动脉起源于单个冠状动脉窦17例(42.50%),为所有畸形中比率最高者,其中第一位为1LCx1R,占25%,第二位为2LCx2R,占17.50%,与其他类型的发病率比较差异有统计学意义(P<0.05)。与术中所见对比,164例患者以患者为单位的经胸超声心动图诊断敏感性、特异性、阳性预测值及阴性预测值分别达90.0% , 95.2%, 85.7%及96.7%。以手术结果为对照,53例术前双源CT及经胸超声心动图对冠状动脉畸形诊断以患者为单位的诊断准确性相似,差异无统计学意义(P值均>0.05),其中双源CT与经胸超声心动图敏感性为94.7%对94.7%、特异性为94.1%对91.7%、阳性预测值为90.0%对85.7%,阴性预测值为96.7%对96.9%。结论:1超声心动图能够在ASO术前准确诊断绝大部分TGA患者的冠状动脉解剖类型,与双源CT具有相似的诊断准确性,可替代双源CT作术前检测。2本研究共发现10种异常冠状动脉类型,有助进一步认识TGA冠状动脉畸形的发生规律。更多还原

【Abstract】 OBJECTIVE Dual-source computed tomography(DSCT),3.0T contrast enhanced whole-heart coronary magnetic resonance angiography(whole-heart CE-CMRA)and transthoracic doppler echocardiography ( TTE) are emerging non-invasive modalities to detect coronary stenosis.Yet,to our knowledge,there has been no prospective ,simultaneous comparison of these three imaging approaches for detecting significant coronary stenosis..Therefore , we conducted a study to determine the diagnostic accuracy of DSCT,3.0T whole-heart CE-CMRA and TTE for the detection significant coronary stenosis.METHODS A prospective,simultaneous and blind study was performed in 51 consecutive patients with suspected or known coronary artery diseases who underwent DSCT、3.0T whole-heart CE-CMRA and TTE within one day before invasive coronary angiography. The diagnostic accuracy of the 3 modalities for detecting significant coronary stenosis (≥50% luminal diameter stenosis) was compared blindly with quantitative invasive coronary angiography as the reference standard.RESULTS 1. DSCT had similar interpretable segments (486/512,96.0%) as 3.0T whole-heart CE-CMRA(479/512,93.6%),higher than TTE did(398/512,77.7%) 2.According to the quantitative coronary angiography,on a per-vessel basis,DSCT and 3.0T whole-heart CE-CMRA had similar sensitivity (94% vs. 88%, p>0.05), specificity (90% vs. 88%, p>0.05), PPV (82%, vs. 78%, p>0.05) and NPV (97%, vs. 95%, p>0.05) for detection of≥50% coronary stenosis, although 3.0T whole-heart CE-CMRA showed a slight trend of inferiority. TTE had significantly lower sensitivity(65%, p<0.05) ,specificity ( 72% ,p<0.05) ,PPV(52%,p<0.05) and NPV(81%,p<0.05) for the detection of≥50% coronary stenosis as compared with DSCT and 3.0T whole-heart CE-CMRA. 3. On a per-segment basis, DSCT and 3.0T whole-heart CE-CMRA had similar sensitivity (95% vs. 89%, p>0.05), specificity (90% vs. 90%, p>0.05), PPV (86%, vs. 85%, p>0.05) and NPV (97%, vs. 93%, p>0.05) for detection of≥50% LM +LAD stenosis, DSCT and 3.0T whole-heart CE-CMRA had similar sensitivity (95% vs. 85%, p>0.05),specificity (95% vs. 92%, p>0.05), PPV (90%, vs. 79%, p>0.05) and NPV (97%, vs. 95%, p>0.05) for detection of≥50% Cx stenosis . DSCT and 3.0T whole-heart CE-CMRA had similar sensitivity (94% vs. 88%, p>0.05),specificity (85% vs. 82%, p>0.05), PPV (70%, vs. 71%, p>0.05) and NPV (97%, vs. 93%, p>0.05) for detection of≥50% RCA stenosis. As compared with DSCT and 3.0T whole-heart CE-CMRA,TTE had significantly lower sensitivity(79%, p<0.05) ,specificity(78% ,p<0.05) , PPV(68%,p<0.05) and NPV(86%,p<0.05) for the detection of≥50% LM +LAD stenosis; significantly lower sensitivity(46%, p<0.05) ,specificity ( 68% ,p<0.05) , PPV(33%,p<0.05) and NPV(79%,p<0.05) for the detection of≥50% Cx stenosis; and significantly lower sensitivity(65%, p<0.05) ,specificity ( 69% ,p<0.05) ,PPV (52%,p<0.05) and NPV(79%,p<0.05) for the detection of≥50% RCA stenosis. TTE had higher sensitivity(79%)for the detection of≥50% LM +LAD stenosis than that of RCA(65%)and Cx(46%) . 4. By echocardiography,stenotic max diastolic velocity (MDV) for LAD was 3.6±1.9M/S,prestenotic to stenotic MDV ratio was 3.3±1.2;MDV for Cx was 3.0±0.5M/S,prestenotic to stenotic MDV ratio was 2.8±0.6;MDV for Cx was 2.8±0.6M/S,prestenotic to stenotic MDV ratio was 2.7±0.3. There were no significant differences in the MDV, and no significant differences in the prestenotic to stenotic MDV among LAD,Cx and RCA .CONCLUSIONS Visual assessment of coronary diameter stenosis severity by DSCT or 3.0T whole-heart CE-CMRA allows identification of significant(≥50%)coronary stenosis with a similar high diagnostic accuracy.Although DSCT showed slightly superior to 3.0T whole-heart CE-CMRA ,there was no significant difference. Because of the lower success rate and less number of interpretable segments, TTE performed worse than DSCT or 3.0T whole-heart. 3.0T whole-heart CE-CMRA permits reliable noninvasive detection of significant coronary stenosis without the use of radiation and potentially hazard iodine contrast agent. TTE can evaluate resting coronary flow dynamics by detection of stenotic coronary artery velocity ,while it is portable and not expensive. TTE can be a helpful supplement to DSCT and 3.0T whole-heart CE-CMRA. OBJECTIVE To identify the transmural extent of myocardial infarction (TME) is critical for making decision of revascularization and evaluating prognosis. However,it is not clear whether 2D-speckle tracking imaging(2D-STI) should be preferred in chronic myocardial infarction to estimate infarct size. In addition, the application method and value of 2D-STI is unclear.The objective of the present study was to investigate the feasibility of 2D-STI to evaluate the transmural myocardial infarction .METHODS A randomized,simultaneous and blind study was performed in 43 consecutive chronic myocardial infarction patients who underwent 2D-STI、delayed-enhanced magnetic resonance imaging ( DE-MRI ) and biochemical examination. On the global level of left ventricle,global longitudinal strain(GLS),rotation and torsion were separately measured. On the segmental level of left ventricle,segmental longitudinal,circumferencial and radial strain, rotation angle as well as diastolic wall thickness were separately analyzed.Global infart size was calculated as infarct volume( a percentage of total myocardial volume )and infart masses, Segmental transmurality was calculated in a 18-segment LV model as infarct volume divided by myocardial volume per segment. Segments with≥50% contrast enhancement were judged transmurally infracted, <50% contrast enhancement were judged non-transmurally infracted. Segments were divided into transmural myocardial infarction(TMI)group and non-transmural myocardial infarction(NTMI) group.Biochemical indices were detected meanwhile including high sensitive c-reactive protein、NT Pro BNP and homocysteine.RESULTS①On the global level, GLS significantly correlated with infarct volume (P=0.008),while the correlation coefficient was 0.620. GLS and maximal apical rotation rate were significant predictors of infarct volume (P=0.005,0.014), while the Beta was 0.720 and 0.592 respectively; GLS was significant predictors of infarct masses (P=0.024), while the Beta was 0.545.The biochemical indices mentioned above were not significant predictors for infarct volume and masses;②On the segmental level,compared with non-transmural infarct group,diastolic wall thickness ,SR(ES), SC(ES), SR Peak G, SL(ES), SC Peak, SL Peak G, SL Peak and SC Peak G (ranked in descending order according to the area under ROC curve, the same below) decreased significantly in apical anterior segments(P<0.05);SC(ES),SC Peak G,SC Peak and SL(ES) decreased significantly in apical lateral segments(P<0.05);SL Peak ,SL Peak G, SC Peak ,SL(ES),SC(ES) and SR(ES) decreased significantly in apical anteroseptal segments(P<0.05);SR Peak G,SR(ES)及SL(ES) decreased significantly in apical posterior segments(P<0.01);SC Peak G,SC Peak及SC(ES) decreased significantly in apical inferior segments(P<0.01);diastolic wall thickness decreased significantly in apical septal segments(P=0.007) ;there was no significant different in rotation angle between two groups.③ROC analysis showed diastolic wall thickness had the best ability to identify transmural infarction both in apical anterior and septal segments,using a cut-off of 5.5MM,diastolic wall thickness had a sensitivity of 83.3% and specificity of 85.7% for apical anterior segments;while a sensitivity of 100% and specificity of 61.5% for apical septal segments. SC(ES)had the best ability to identify transmural infarction in apical lateral segments,using a cut-off of -5.4650,SC(ES)had a sensitivity of 83.3% and specificity of 85.7%. SR Peak G had the best ability to identify transmural infarction in apical posterior segments,using a cut-off of 12.265,SR Peak G had a sensitivity of 100% and specificity of 80.0%,Both SC Peak G and SC Peak S had the best ability to identify transmural infarction in apical inferior segments,using a cut-off of -5.3200,SC Peak G and SC Peak S had a sensitivity of 100% and specificity of 88.8%.CONCLUSIONS①On the global level,GLS correlated significantly with global infarct size.Apical rotation rate is better than torsion angle and torsion rate in predicting infarct size.②On the segmental level , longitudinal strain ,circumferencial strain and radial strain discriminated between non-transmural and transmural infarction in different apical segment respectively.The discriminating ability was different for different strain indices,which may reflect the complexity of heterogeneity in myocardial fibers direction and transmural strain of different apical segments. OBJECTIVE Arterial switch operation (ASO) is the anatomical correction procedure of transposition of great arteries (d-TGA).Accurate preoperative identification of anomalous coronary artery anatomy pattern is critical for the success of ASO. We conducted the study to evaluate the diagnostic value on defining coronary artery anatomy by transthoracic echocardiography(TTE) and dual- source CT(DSCT),and to conclude how to improve the preoperative diagnostic accuracy of coronary artery anatomy by TTE.METHODS 164 patients underwent TTE before ASO. Diagnostic accuracy of TTE was evaluated using surgical diagnosis as a reference. 49 patients had TGA with intact ventricular septum (TGA/IVS),77 patients had TGA with ventricular septum defect(TGA/VSD),38 patients had double outlet right vetricle with TGA(DORV/TGA).Among 164 patients,there were 53 consecutive patients underwent both TTE and DSCT within 3 days before ASO. A blind study was performed using surgical diagnosis as a reference.RESULTS According to surgical diagnosis ,among 164 patents,there were 124 case(s75.61%) with normal coronary artery,and 40 case(s24.39%) with anomalous coronary arteries,The incidences of anomalous coronary artery were significantly higher in TGA/VSD combined with DORV/TGA than that in TGA/IVS ( 80.00%, 32 cases versus 20.00%,8 cases,P<0.05). There were 10 different anomalous coronary patterns found in this study.The incidence of the anatomical pattern with all coronary arteries originated from one coronary sinus was the highest(17 cases,42.5%).Among them, 1LCx1R ranked the first(25%), 2LCx2R ranked the second(17.5%).The incidence was significant different as compared with other anomalous patterns(P<0.05). According to surgical diagnosis, in the patient-based analysis, TTE diagnostic sensitivity, specificity, PPV, NPV in 164 patients were 90.0% , 95.2%, 85.7%及96.7% respectively. On a patient basis, DSCT and TTE had similar sensitivity ( 94.7% versus 94.7%, P>0.05) , specificity (94.1% versus 91.7%, P>0.05) , PPV ( 90.0% versus 85.7%, P>0.05) and NPV(96.7% versus 96.9% , P>0.05) for detection of coronary anatomical pattern in 53 cases.CONCLUSIONS 1. Coronary anatomical pattern in TGA could be accurately by TTE for most of cases. TTE and DSCT had similar diagnostic accuracy.TTE could be used as a preoperative diagnostive modality in stead of DSCT.There were 10 different types of anomalous coronary patterns found in this study ,which could benefit understanding the pathogenesis of anomalous coronary pattern in TGA更多还原