【作者】 刘红艳；

【导师】 张雪林；

【作者基本信息】 南方医科大学 ， 影像医学与核医学， 2009， 博士

【摘要】 研究目的1.探讨64层螺旋CT（MSCT）冠状动脉成像质量的影响因素。2.以常规冠状动脉造影（CAG）为对照,评价64层MSCT诊断冠脉狭窄≥50%的诊断价值;评价斑块的性质与狭窄程度的关系。3.探讨64层MSCT在冠状动脉变异中的临床应用价值。4.探讨64层MSCT冠脉成像（CTCA）在满足临床诊断的图像质量下,通过调整不同个体的管电流（mA）获得一致图像噪声,进而降低辐射剂量的方法。材料与方法1.64层MSCT冠状动脉成像质量的影响因素1.1一般资料93例行64层MSCT冠状动脉成像,男64例,女29例,年龄35～82岁,平均57.30±9.12岁。心前区不适48例,冠心病病史25例,体检20例。受检者排除明显心律不齐、呼吸屏气不良、无法配合检查、心肾功能不全及碘对比剂过敏者,增强扫描均签署知情同意书。1.2仪器与方法使用GE Light Speed 64 VCT,采用回顾性心电门控技术进行冠状动脉成像;心电监护仪使用美国IVY 3150型ECG;使用美国LF CT9000双筒型高压注射器及350mgI/ml或370mgI/ml非离子型对比剂。扫描参数及方法:120 KV,650～680 mA,0.35 s/r,64×0.625 mm,视野25 cm,矩阵512×512,螺距0.18～0.20。先行探测循环时间扫描（TBS）测定延迟时间,用18G套管针在肘正中静脉以5ml/s的流率团注20ml对比剂,随后注入20ml生理盐水,测定臂心循环时间及最佳扫描延迟时间。继而行冠状动脉扫描,扫描范围从气管隆突至心脏膈面,经肘静脉以5ml/s的流率注入对比剂约60～80ml,由已确定的延迟时间（18～26s）触发扫描,当对比剂注射完后以相同的流速注入40ml生理盐水。扫描时间约为6～8s。93例中73例采用常规屏气后直接扫描,20例采用屏气5～8s后扫描。心率＞75次/min时,口服倍他乐克12.5～25mg以降低心率。根据患者心率不同,采用不同的扇区重组方法:心率＜75次/min时采用单扇区重组,心率≥75次/min时采用多扇区重组。1.3图像处理与重组使用AW 4.2后处理工作站,先取75%R-R间期图像进行重组,图像不佳者,重建45%R-R时相图像,若仍有伪影,再重建55%或65%R-R时相图像。选择不同的成像方法如容积成像（VR）、多平面重组（MPR）、曲面重组（CPR）及最大密度投影（MIP）等,并结合横断面图像综合判定。1.4冠状动脉分段采用改良10段分法:右冠状动脉（RCA）分为近段（R1）、中段（R2）、远段（R3）;左主干（LM）;前降支（LAD）分为近段（L1）、中段（L2）、远段（L3）、对角支（L4）;回旋支（LCX）分为近段（C1）、远段（C2）。1.5图像质量评价以血管段为单位的图像质量判定标准:优-图像无伪影,血管边缘清晰、连续且充盈良好;良-图像周围有少量伪影,血管边缘欠光滑,有轻微阶梯和错层伪影,但可以做出诊断;差-图像显示模糊,伪影明显,血管不连续、充盈差,经多期相重组依然无法进行冠脉分析诊断。每一患者冠脉图像质量评分标准:5分-全部血管段图像质量均为优;4分-有一个血管段图像质量为良;3分-有两个血管段图像质量为良;2分-三个或三个以上血管段图像质量为良;1分-有一个或一个以上血管段图像质量为差。由三位主治以上的医师对冠状动脉图像质量进行评价,意见不一致时采用多数人意见或协商解决。心率分为3组:＜65次/min、65～75次/min、＞75次/min;心率波动次数亦分为3组:＜5次/min、5～9次/min、≥10次/min。在原始横断面图像上对冠状动脉主要血管段（直径≥2mm）进行CT值测量,感兴趣（ROI）设为6mm²。1.6统计学分析使用SPSS 13.0统计软件,对于CTCA图像质量影响因素评价采用多变量有序Logistic回归分析,并采用偏相关及Spearman非参数相关分析;不同心率及心率波动组对图像质量的影响、重建间隔的选择及屏气状态对图像质量的影响等计数资料采用独立样本非参数检验;计量资料比较采用独立样本t检验,P＜0.05认为差异具有统计学意义。2.64层MSCT冠状动脉成像在冠心病中的应用价值2.1一般资料回顾性分析行64层MSCT冠状动脉成像50例,所有患者均在1月内行CAG。男41例,女9例。年龄34岁～82岁,平均57.25±11.50岁。其中冠心病38例、高血压22例、糖尿病11例。2.2仪器与方法CTCA检查同上。CAG检查:使用西门子大型悬吊式C臂机及其配套工作站,对比剂用欧乃派克,采用Judkins法常规股动脉或桡动脉插管,分别行左冠状动脉4～6个标准体位,右冠状动脉2～3个标准体位。2.3狭窄判定标准采用美国心脏协会推荐的冠状动脉15分段法。狭窄程度采用国际上通用的目测直径法,即以血管腔狭窄部位近心端相对正常的管腔直径作为参考值,对其狭窄程度进行定量评价。正常冠状动脉为血管腔光滑,无粥样硬化斑块,前向血流TIMI（Thrombolysisin myocardial infarction）3级。狭窄程度判定:轻度狭窄,血管腔直径狭窄百分数＜50%;中度狭窄,血管腔直径狭窄百分数50%～75%;高度狭窄,血管腔直径狭窄百分数76%～99%;闭塞组为血管腔完全闭塞,前向血流TIMI 0级。64层MSCT与CAG狭窄程度的判定分别有两位有经验的放射科医师采用盲法对冠状动脉进行评价,意见不一致时商议后确定。2.4斑块的分类采用国内通用的CT值测量方法。软斑块:CT值＜60HU（部分可为负值）;纤维性斑块:CT值60～129HU;钙化斑块:CT值≥130 HU;混合性斑块:包括钙化、非钙化成分。2.5统计学方法使用SPSS 13.0统计软件,以CAG为标准,计算64层MSCT诊断临床有意义的冠状动脉狭窄（≥50%）的准确性、灵敏度（Se）、特异度（Sp）、阳性预测值（PPV）、阴性预测值（NPV）;配对资料采用配对X²检验,等级资料采用非参数检验及Bonferroni多重比较。P＜0.05认为差异有统计学意义。3.64层MSCT冠状动脉成像对冠状动脉变异的诊断价值3.1一般资料收集2006年2月至2008年12月1800例行64层MSCT冠状动脉成像的临床资料,发现先天性冠状动脉变异47例。男38例,女9例,年龄37～72岁,平均51.35±10.34岁。29例偶感胸痛、胸闷,5例临床诊断为冠心病、心绞痛,13例为健康体检者。64层MSCT发现冠状动脉粥样硬化7例,其中有临床意义的冠状动脉狭窄（≥50%）4例。2例分别在MSCT检查后半月内行CAG,1例CAG未寻找到冠脉开口,故行64层MSCT冠脉成像。3.2仪器与方法:CTCA及CAG检查方法同上。4.64层MSCT冠状动脉成像在固定噪声水平低剂量扫描中的临床初探4.1一般材料60例患者分为2组。回顾性分析2008年11月～12月行64层CTCA检查的患者30例作为固定mA组,男22例,女8例,年龄37～72岁,平均49.10±9.04岁;随后采用前瞻性研究方法分析2009年1月～2月行CTCA检查的患者30例作为个体mA组,即根据个体差异改变扫描mA,男24例,女6例,年龄34～70岁,平均49.73±9.92岁。因早搏或呼吸影响成像质量的患者、心肾功能不全及对碘对比剂过敏者不纳入本研究,需控制心率≤70次/min、心率波动＜5次/min。4.2仪器和设备:同上。4.3 CTCA检查方法回顾性研究:30例固定mA组,采用固定扫描参数。探测循环时间（TBS）扫描:120KV,80mA,5mm层厚,标准重建算法;心脏扫描（CA）:120KV,680mA,0.35 s/r,64层×0.625mm,标准重建算法,螺距0.18。选择ROI（1cm×1cm）进行噪声标准差（SD）的测量,分别在TBS、CA于主动脉分出左主干的上下三层面测量图像噪声,取其平均值作为该病人的SD值。评价SD_TB、BMI与SD_CA的相互关系并得出以SD_CA为因变量的线性回归方程;根据噪声和射线剂量的关系,可得到当CA达到能满足临床诊断的噪声（SD₀）时不同个体SD_TB所对应的管电流调制表。前瞻性研究:连续选取30例拟行64层CTCA检查的患者作为个体mA组,根据不同个体TBS扫描时获得的SD_TB值,以及管电流调制表确定CA扫描时的mA值,其它扫描参数同固定mA组。以同法测量图像噪声。余CACT检查方法及后处理技术同第一部分。4.4图像噪声、质量评分由2名医师用5分法评价固定mA组的噪声水平及图像质量,并确定满足临床诊断的图像噪声水平SD₀值,同法对个体mA组图像质量进行评分。评分标准:5分-冠状动脉边缘平滑锐利,分支及远段显示好;4分-冠状动脉显示好,有一定噪声,远段分支显示好;3分-冠状动脉主干显示好,噪声较大,但不影响斑块的观察及诊断;2分-噪声较大,边缘毛糙,远段显示差,勉强诊断;1分-噪声太大,无法诊断。4.5统计学分析使用SPSS 13.0统计软件,对于SD_CA与SD_TB、BMI的关系采用Person相关性分析,并建立以SD_CA为因变量的线性方程;对于SD_CA与图像质量评分的关系采用Spearman相关性检验;对于两组年龄、BMI、图像质量评分、图像噪声、CTDI_vol及ED等的比较采用独立样本t检验。P＜0.05认为有统计学差异。结果1.64层MSCT冠状动脉成像质量的影响因素1.1 730段冠状动脉节段中,可用于评价的血管段占93.7%（684/730）;1.2心率及心率波动是冠状动脉成像质量的主要影响因素,且呈负相关;心率的影响最大（r=-0.422,P=0.000）,其次是心率波动（r=-0.257,P=0.015）;心率≤75次/min、心率波动＜10次/min时冠脉成像质量较好;1.3心率≤75次/min组85.7%病例在65%～75%R-R间期重组即可获得较好CTCA图像质量,心率＞75次/min组47.1%病例需再选择45%～55%R-R间期重组图像进行后处理分析,不同心率组重组间期的选择差异具有统计学意义（Z=-2.841,P=0.004）;1.4以相同流速5ml/s注入对比剂后,心率≤75次/min组较心率＞75次/min组主要血管段CT值高约50HU,仅左主干（LM）的CT值在高、低心率组间有统计学差异（t=2.394,P=0.028）,但两组CT值均能满足临床诊断要求;1.5屏气5～8s后扫描较屏气后直接扫描CTCA成像质量好（Z=-2.571,P=0.01）;但两组图像质量的优良率差异无统计学意义（Z=-0.956,P=0.339）。2.64层MSCT冠状动脉成像在冠心病中的应用价值2.1 64层CTCA检查中,50例共计683个冠脉节段可用于评价,153个节段发现斑块及狭窄性病变;2.2 64层MSCT诊断冠脉节段≥50%狭窄与CAG符合率为84.0%（63/75）,诊断冠脉节段或主要分支血管狭窄≥50%的Se、Sp、PPV、NPV及准确性分别为84.0%、97.9%、81.8%、98.2%、96.5%或87.5%、95.1%、87.5%、95.1%、93.0%;2.3不同斑块性质对冠脉狭窄程度影响不同（X²=30.003,P=0.000）。单纯钙化斑块所致的狭窄程度常较轻,混合斑块所致的狭窄程度往往较重,而非钙化斑块所致的狭窄程度分布无一定特异性,常与CAG表现一致;钙化与非钙化斑块、钙化与混合斑块在狭窄程度的分布上差异有统计学意义（P=0.000）。3.64层MSCT冠状动脉成像对冠状动脉变异的诊断价值1800例CTCA检查中共检出冠状动脉变异47例,占2.6%。冠状动脉开口起源异常32例,其中包括左、右冠状动脉窦上起源12例、多个开口7例,右冠状动脉起源于左冠状窦或窦上嵴10例、左冠状动脉起源于右冠近段1例、左旋支起源于右冠状窦或第一对角支各1例;LCX缺如1例;单一左冠状动脉2例;壁冠状动脉-心肌桥12例。4.64层MSCT冠状动脉成像在固定噪声水平低剂量扫描中的临床初探4.1固定mA组SD_TB与SD_CA相关性较高（r=0.867）,可得出SD_CA（y）与SD_TB（x）的线性方程:y=1.747x+1.920（调整R²=0.736）;4.2固定mA组图像噪声与质量评分相关性分析显示,SD_CA与评分呈负相关（r=-0.412,P=0.024）,当SD₀=28HU可获得满意的图像质量;4.3根据射线剂量与噪声的关系,可计算出获得一致图像噪声SD₀时不同个体所需mA（z）关系式:z={（1.747x+1.920）/28}²×680 mA(x=SD_TB);4.4个体mA组与固定mA组相比,CTDIvol减少了10.64%（t=7.038,P=0.000）、ED减少了15.03%（t=7.038,P=0.000）,但两组图像质量差异无统计学意义（t=0.530,P=0.598）。结论1.心率及心率波动是64层MSCT冠脉成像质量的主要影响因素。严格呼吸屏气训练,控制心率及心率波动,选择合适的重组时相可以提高冠状动脉成像质量。2.64层MSCT对于冠心病的诊断具有较高的特异度及准确性,可作为一种有效的筛查手段,但仍有一定的局限性。3.64层MSCT是一种安全、准确、微创地筛查和诊断冠状动脉变异的重要手段,同时对于CAG的导管入路检查及治疗有一定的指导作用。4.64层MSCT冠脉成像时,对于不同体质采用不同的曝光剂量,可在保证图像质量不变的前提下,有效的减少X线辐射剂量。更多还原

【Abstract】 Objective1.To explore the main influence factors on coronary artery image quality with 64 multi-slice spiral CT（MSCT）.2.To study the value of 64 MSCT coronary angiography in diagnosis of coronary stenoses（≥50%）in patients with coronary heart disease（CHD）,Compared with conventional angiography（CAG）,and to evaluate the correlation of the plaque and the stenoses.3.To discuss the clinical application value of 64 MSCT in the diagnosis of congenital coronary artery anomalies.4.To evaluate the different individual adapted tube current selection method for obtaining consistent image noise and reducing radiation dose for patient population on 64-slice spiral CT coronary angiography（CTCA）.Material and Method1.The influence factors on coronary artery image quality with 64 MSCT1.1 General material93 cases had 64 MSCT scan for coronary angiography.There were 64 male,29 female,and average age 57.30±9.12 years old.Clinical symptom included precordia complaint（n=48）,history of CHD（n=25）,and health examination （n=20）.Subjects were ruled out by obvious irregularity of cardiac rhythm,worse breathing,unability to match the exam,heart failure,renal inadequacy and iodi hypersensitiveness.Every one had to have enhanced CT signed consent.1.2 Main instrument or equipmentRetrospective electrocardiogram（ECG）-gated coronary CT angiography was performed with GE Light Speed 64 VCT.ECG monitor with type of IVY 3150 was made in America.The type of two-bucket high pressure syringe made in America was LF CT 9000 and non-ion contrast medium with 350mgI/ml or 370mgI/ml were used.1.3 Examination technique and image processingScanning parameter:120 KV,650～680 mA,0.35 s/r,64×0.625 mm,FOV=25 cm,matrix=512×512,pitch=0.18～0.20,standard reestablish algorithm.Firstly,a timing bolus scan（TBS）was obtained at the level of the aortic root with the administration of 20ml non-ion iodic contrast medium（350mgI/ml or 370mgI/ml） followed by 20ml saline solution at 5ml/s.So transit time and optimun scanning time was able to decide.Subsequently,a bolus of 60～80 ml of contrast medium followed by 40ml of saline solution was injected at the same rate.The scanning range covered the heart from the level of tracheal bifurcation to the diaphragm.The total scanning time was about 6～8s.All of 93 cases,73 cases made a direct scan after breathholding and 20 cases done a scan after 5～8s breathholding.A dose of 12.5～25mg of a adrenergic blocking agent,metoprolol was administered orally before CT examination if the patient’s heart rate was more than 75bpm.Different recombination algorithm were selected according to different heart rate,such as heart rate＜75bpm with mono-recombination or heart rate≥75 bpm with multi-recombination algorithm.AW4.2 postprocession workstation was used.Raw CT datas were firstly reconstructed at 75%R-R of the cardiac cycle.If the image qulity was not better, 45%reconstruction phase was needed.55%or 65%reconstruction phase was also required if artifact was still present.Different processes such as volume rendering （VR）,multi-planar reformation（MPR）,curved planar reformation（CPR）and maximum intensity projection（MIP）were used to assess for image quality combining cross section image.1.4 Coronary artery segmentTen coronary segments were analyzed in each patient according to the way of improvement.The right coronary artery（RCA）included proximal segment（R1）, middle segment（R2）,distal segment（R3）.The left main artery（LM）diverged two main segements including of left anterior descending artery（LAD）and left circumflex artery（LCX）.The LAD contained proximal segment（L1）,middle segment （L2）,distal segment（L3）and diagonal segment（L4）.The LCX consisted of proximal segment（C1）,distal segment（C2）.1.5 Image quality evaluationThe image quality based on segment was assessed by rated as excellent,reduced but still diagnostic and non-diagnostic:no artifact,a sharp blood vessel with good continuous engorge;mild artifact,the vessel with no more smooth edge,but assessable;severe artifact,vague image,vessel discontinuation,less engorge, unassessable for the vessel by different reconstruction phase.The image quality based on every patient was defined by 5-point as follows: 5=all excellent image quality for all segments;4=only one segment with fine or diagnostic image quality;3=two segments with fine or diagnostic image quality; 2=equal or more than three segments with fine or diagnostic image quality;l=one or more than one segment with non-diagnostic.The overall quality for all images was assessed blinded and randomised by three experienced radiologists in a single consensus reading.Major people’s opinion and negotiation for result may be used if opinion was on discrepancy.Patients were divided into 3 groups with heart rates:less than 65 bpm;between 65～75 bpm;more than 75 bpm.Three groups were also divided according to heart wave:less than 5 bpm; between 5～9 bpm;more than or equal 10 bpm.The CT value of main coronary segment（diameter≥2mm）was measured on primary cross section with ROI=6mm².1.6 Statistical analysisThe SPSS 13.0 software was applied in this study.Multiple variable Logistic regression was used to evaluate the influence factors on image quality.The spearman’s and partial bivariate correlation were performed to analyze the different influence factors on image quality.The independent sample non-parameter test was applied to evalute the influence of heart rate,heart wave,reconstruction phase and breathholding on image quality.The independent sample t test was used to compare the mean datas.A P value of less than 0.05 was considered significant.2.The clinical application of 64 MSCT coronary angiography in patients with CHD2.1 General material50 cases with both 64 CTCA scan and CAG exam in one month for coronary angiography were retrospective analyzed.There were 41 male,9 female,and average age 57.25±11.50 years old.Of all patients,there were 38 cases with history of CHD, 22 cases with history of high blood pressure,and 11 cases with history of diabetes.2.2 Main instrument and methodThe method of CTCA was same as Part 1.CAG was performed with Siemens macro-type suspled C-arm and its matched workstation.Constrast medium was used with Omnipaque.The conventional cannula of arteria femoralis or arteria radialis was applied with Judkins,including 4～6 standard podition of left coronary artery（LCA）and 2～3 standard position of RCA.2.3 Stenoses assessment 15 coronary segments were analyzed in each patient according to the fractionation method recommended by America heart institution.The extent of coronary stenoses was assessed with international used eye measurement.Normal coronary artery was a smooth blood vessel,no atherosclerotic plaque,grade 3 with TIMI（thrombolysis in myocardial infarction）.The stenosed extent of blood vessel diameter was ranked 4 groups:light stenoses,less than 50%; middle stenoses,50%～75%;weight stenoses,76%～99%;obstruction,blunting lumen of blood vessel and grade 0 with TIMI.The overall stenosed extent on 64 CTCA and CAG was respectively assessed blinded and randomised by two experienced radiologists in a single consensus reading.Negotiation for result may be used if opinion was on discrepancy.2.4 Plaque rankPlaque was ranked into 4 groups by current measurement of CT value:soft plaque,less than 60HU（partly negative value）;fiber plaque,between 60～129HU; calcified plaque,more than 130 HU;mixed plaque,including calcifed and noncalcified component.2.5 Statistical analysisThe SPSS 13.0 software was applied in this study.The accurance,sensitivity（Se）, specificity（Sp）,positive predictive value（PPV）,negative predictive value（NPV）were all calculated to evaluate the diagnosis of coronary stenoses（≥50%）on 64 CTCA. The matched-pair x² test,independent sample non-parameter test or Bonferroni mutiple comparison was applied to evalute the pairing data or ranked data.A P value of less than 0.05 was considered significant.3.The value of 64 MSCT in the diagnosis of coronary artery anomalies.3.1 General material64 MSCT coronary artery angiography datas of 1800 patients between February 2006 and December 2008 were analyzed retrospectively to find the coronary artery anomalies in 47 patients.There were 38 male,9 female,age from 37 to 72 years old, and average age 51.35±10.34 years old.Clinical symptom included occasional chest pain or distress（n=29）,clinical diagnosis of CHD or angina（n=5）,and health examination（n=13）.7 cases of coronary atherosclerosis were revealed using 64 MSCT,including 4 cases of clinical coronary stenosis（≥50%）.Two cases had CAG exam in half past one month after 64 MSCT coronary angiography.One case had 64 CTCA scan because of no detecting coronary gab on CAG.3.2 Main instrument and methodThe method of CTCA and CAG was same as Part 2.4.The way of low dose scanning on fixed noise level in 64 MSCT cardiac imaging4.1 General materialTotal 60 cases were divided into two groups in this study.Firstly,30 cases as fixed-mA group were retrospectively analyzed with CTCA exam from Nov to Dec in 2008.Among them,there were 22 male,8 female,and average age 49.10±9.04 years old.Sequently,30 cases of individual-mA group were prospectively studied on CTCA scan from Jan to Feb in 2009.Among them,there were 24 male,6 female,and average age 49.73±9.92 years old.The patients were ruled out because of bearing premature,worse breathing,heart failure,renal inadequacy and iodi hypersensitiveness. All subjects must keep heart rate≤70 bpm and heart wave＜5 bpm.4.2 Main instrument or equipment:same as Part 1.4.3 Examination technique and image processingRetrospective study:We firstly analyzed 30 patients underwent CTCA using 64 MSCT with standard scan protocol（TBS:120KV,80mA,5mm thick,standard reestablish algorithm;CA:120KV,680mA,0.35s/r,64×0.625 mm,pitch=0.18, standard reestablish algorithm）to establish the relationship between SD_TB、BMI and SD_CA.An excel table was established to predict the required mA to achieve a desired SD₀ for patient with single SD_TB.The image noise was measured for each patient using the average of three consecutive slices in the ascending aorta with region of interest（ROI）cursor of 1 cm×1 cm.Prospective study:We then scanned 30 patients with individual SD_TB-adapted mA from the table to evaluate the robustness and practicability of this method.We did not use other dose reduction techniques in this study.The way of SD mesurement was same as fixed mA group.CT dose index volume（CTDIvol）and effective dose （ED）were recorded.The other examination technique and image processing were same as Part 1.4.4 Image noise and quality evaluationThe overall quality for all images was assessed blinded and randomised by two experienced radiologists in a single consensus reading based on a five point grading scale as follows:5=clear delineation of small structures,distinct anatomic detail and sharp vessel;4=clear anatomic detail with mild increase in image noise;3=distinct increase in image noise with still unaffected diagnostic image quality;2=obscured anatomic detail due to deterioration in image quality,extensive blurring and distinct increase in image noise leading to unsure diagnosis or even resulting in an insufficient evaluation of diagnosis;and l=non-diagnostic.4.5 Statistical analysisThe SPSS 13.0 software was applied in this study.The Pearson’s bivariate correlation was used to analyze the relationship of the SD_TB,BMI and SD_CA,and to establish a equation of linear regression.The Spearman rank correlation was applied to evalute the relationship of SD_CAand image quality analysis.The mean datas was presented with（?）±S pattern.The independent sample t test was used to compare the values of the two groups and statistical difference was granted as P value＜0.05. Results1.The influence factors on coronary artery image quality with 64 MSCT1.1 Of 730 segments,the rate of segment used to estimate was 93.7%（684/730）.1.2 Heart rate and heart wave were the main influence factors on coronary artery image quality.The relationship of them showed negative correlation,with heart rate closer to image quality（r=-0.422,P=0.000）,and then the heart wave（r=-0.257, P=0.015）.The better image quality was obtained when the heart rate was≤75 bmp, the heart fluctuation was＜10 bmp.1.3 87.5%cases of all only selected 65%～75%reconstruction phase to get better image quality if the heart rate was≤75 bpm.When the heart rate was＞75bpm, 45%～55%reconstruction phase was reselected by 47.1%cases of all.The selected reconstruction phase of different heart rate groups had a statistical significance （Z=-2.841,P=0.004）.1.4 The CT Attenuation values of the main coronary segment on the heart rate≤75bpm group were higher by 50HU than the heart rate＞75 bpm group.Only CT value of the LM in two groups had a statistical significance（t=2.394,P=0.028）,but all satisfied clinical diagnosis.1.5 The excellent rate of image quality on the scan after 5⁸s breathholding is higher than the direct scan after breathholding（Z=-2.571,P=0.01）,but the rate as excellent and reduced but still diagnostic of the two groups had no significant difference（Z=-0.956,P=0.339）.2.The clinical application of 64 MSCT coronary angiography in patients with CHD2.1 Total 683 segments of 50 patients were able to evaluate on 64 CTCA.The plaque and stenoses were detected in 153 segments of coronary arteries.2.2 The distribution and extent of 63 coronary arterial segments stenoses（≥50%）revealed by 64 CTCA were correlated exactly with CAG in 75 coronary arterial segment stenoses.The coincidence rate was 84.0%.The diagnosis value of coronary arterial segments or branches on 64 CTCA had little difference,and the Se,Sp,PPV, NPV and accuracy were 84.0%,97.9%,81.8%,98.2%,96.5%or 87.5%,95.1%, 87.5%,95.1%,93.0%respectively.2.3 The different stenosed extent resulted from the different coronary plaque （X²=30.00,P=0.000）.Calcified plaque always caused mild lumen diameter stenoses. However,the mixed plaque aways leaded to serious lumen diameter stenoses.The distribution of stenoses caused by noncalcified plaque was no specificity,and had a same appearance on CAG.The distribution of stenoses between calcified and noncalcified plaque or between calcified and mixed plaque had significant difference （P=0.000）.3.The value of 64 MSCT in the diagnosis of coronary artery anomalies.In consecutive 1800 patients,64 MSCT identified 47 patients by 2.6%with an anomalous coronary artery.Anomalous origin of coronary artery contained 32 cases, including LCA or RCA origin from supra-sinus（n=12）,multi-orifice（n=7）,RCA origin from left coronary artery sinus or sinus crista（n=10）,LCA origin from the proximal segment of RCA（n=1）,LCX origin from right coronary sinus（n=1）or the first diagonal artery（n=1）.The others were absence of LCX（n=1）,single LCA（n=2）, and myocardial bridge（n=12）.4.The way of low dose scanning on fixed noise level in 64 MSCT cardiac imaging4.1 The relationship of SD_TB（x）and SD_CA（y）was closer（r=0.867）to be able to establish a equation of linear regression in the fixed mA group: y=1.747x+1.920（adjusted R²=0.736）4.2 The relationship of SD_CAand image quality analysis showed negative correlation （r=-0.412,P=0.024）.A cardic image noise level（SD₀）of 28 HU was found to be adequate for clinical diagnosis purpose based on the image quality analysis. 4.3 The formula to predict the needed mA（z）for obtaining consistent image noise （SD₀）was to established on the base of SD_TB（x）: z={（1.747x+1.920）/28}²×680 mA.4.4 The t test indicated that the individual mA group for adapting mA method produced much smaller radiation dose for CTDIvol（t=7.038,P=0.000）and ED （t=7.038,P=0.000）than the fixed mA group.And the mean reduced dose rate was 10.64%,15.03%respectively.However,the image quality analysis between the two groups showed no significant difference（t=0.530,P=0.598）.Conclusion1.Heart rate and heart wave are the main influence factors on coronary artery image quality with 64 MSCT.Strictly trainning for breathholding,controlling the heart rate and heart wave,selecting optimal reconstruction phase can elevate the image quality of CTCA.2.64 MSCT has a higher specificity and accuracy in diagnosis of CHD and can be acted as effective screening method.But it also has some limit.3.64 MSCT is a safe,noninvasive,accurate method for screening and diagnosis of congenital coronary artery anomalies,and also a direction for catheter approach on CAG or treatment.4.The use of individual adapted mA selection method is robust and practical to obtain consistent image quality for different patients and may provide dose reduction for smaller patients compared to the use of fixed mA.更多还原