【作者】 王荣品；

【导师】 梁长虹；

【作者基本信息】 南方医科大学 ， 影像医学与核医学， 2010， 博士

【摘要】 目的：1.通过流体模型,检验3.0 T磁共振相位对比法成像(3.0 T PC-MRI)速度编码值(Venc)、编码方向的选择及其对恒定流体流速和变速流体流速测量的准确性和稳定性。2.探讨非屏气电影法用于不能配合屏气患者心功能检查的可行性及评价非屏气多时相采集电影回放法测量左心室功能的价值。3.采用相位对比法磁共振成像(PC-MRI)测量肺、体循环大血管血流动力学信息,探讨肺、体循环差异,同时验证PC-MRI对人体血流参数测量的准确性。4.探讨3.0 T MRI SE序列结合电影成像评价复杂型先天性心脏病(CCHD)双向Glenn分流术(BGS)后心内畸形、主心室功能和房室瓣功能的诊断价值。5.应用PC-MRI结合CE-MRI评估BGS后,肺血流量与肺血管发育关系,进而计算主-肺侧支血管血流量(APCF)和心内分流量。材料与方法：1.固定流体流速和Venc,检验不同编码方向的流体信息；固定流体流速和编码方向,检验不同Venc对固定流速的影响：固定Venc和编码方向,检验PC-MRI对不同恒定流速(注射流率0-5 ml/s)所测流速值与实际流速的差别；检验PC-MRI对变速流体测量的准确性；采用t检验以检验PC-MRI所测流速值与实际流速的差异是否具有统计学意义。2.采用非屏气法和多次屏气法快速稳态平衡进动序列(FIESTA)对15例健康志愿者进行3.0 T MRI检查,用Report Card软件分析左心室功能,比较两种方法测量所得心功能指标,包括左心室舒张末期容积(EDV)、收缩末期容积(ESV)、射血分数(EF)、平均心肌质量(MM)、心输出量(CO)及相应的体表面积标准化的左心室舒张来期容积指数(EDVI)、心肌质量指数(MMI)、心脏指数(CI),采用配对样本t检验分析两种方法测量结果的差异性和相关性。3.使用GE 3.0 T MRI扫描仪,采用PC-MRI对15例健康志愿者的主肺动脉、右肺动脉、左肺动脉、主动脉、上腔静脉和下腔静脉进行血流测量,然后,计算各血管一个心动周期的净血流量和返流分数。采用配对样本t检验分析肺、体血流差异。4.采用Triple-IR和快速稳态平衡进动序列(FIESTA)对22例BGS术后患者进行心脏检查,用Report Card软件分析心内畸形、主心室和房室瓣收缩功能；分析3.0 T MRI测量主心室舒张末期容积(EDV)、收缩末期容积(ESV)、射血分数(EF)及轴缩短率(FS)与超声心动图(UCG)测量相应指标的相关性；应用独立样本t检验分析BGS术后主心室舒张末期容积指数(EDVI)、收缩末期容积指数(ESVI)、心脏指数(CI)及EF与健康志愿者左心室相应功能指标的差异；应用Spearman等级相关对电影MRI所测的房室瓣返流程度与UCG进行检验分析。5.应用3.0 T PC-MRI对22例BGS术后患者肺、体循环各大血管进行血流测量,用Report Card软件计算主动脉搏出量(Qs),上、下腔静脉回流量(Qv),左、右肺动脉血流量(Qp); APCF=Qs-Qv;心内分流量=Qv-Qp。应用CE-MRI测量肺血管发育指标。对肺、体循环血流差异及双肺动脉血流差异采用配对样本t检验；对肺血管发育指标与肺血流量关系采用Pearson’s和Spearman’s相关分析；对不同前向性血流组肺血流量及APCF的差异采用独立样本t检验；MRI与UCG对腔-肺吻合口测量的差异采用配对样本t检验和Pearson’s相关分析。将心内分流量与主心室舒张末期容积指数(EDVI)和UCG测量的房室瓣返流面积进行相关性分析。所有结果取P<0.05为差异有统计学意义。结果：1.只有选择层面(SLICE)或上、下(SI)编码方向才能正确反映质子流动的方向；在Venc适当大于实际流速时,不同注射流率PC-MRI所测流速值与流速真实值之间差异无统计学意义(t=-0.861,P=0.405),二者呈显著正相关关系(r=0.999,P<0.001)；3.0 T PC-MRI在测量变速流体的最大、最小和平均值与相应的实际流速一致性良好。2.非屏气法和多次屏气测得左心室各项功能指标分别为：EDV为(123.85±19.48)ml和(121.97±17.53)ml,ESV为(46.64±8.34)ml和(45.57±9.18)ml,EF为(62.65±5.12)%和(62.39±4.67)%,MM为(106.25±18.07)g和(106.25±18.07)g；EDVI为(70.44±7.11)ml/m2和(69.58±6.53) ml/m2,MMI为(60.83±5.45)g/m2和(60.17±6.34)g/m2,CO为(5.70±0.778)L/min和(5.69±0.88)L/min,CI为(3.26±0.24)L/min/m2和(3.27±0.30)L/min/m2。两种方法所得的上述各项指标经配对样本t检验,差异无统计学显著性意义(P值均>0.05),相关性良好(r=0.633-0.957)。3.主肺动脉、右肺动脉和左肺动脉一个心动周期的平均血流量分别为(69.42±12.35)mL,(38.78±6.81)mL和(32.27±6.49)mL,右肺动脉血流量显著大于左肺动脉血流量(t=3.092,P=0.004),左肺动脉返流分数明显大于右肺动脉返流分数(t=5.502,P=0.001)。主动脉一个心动周期的平均血流量为(70.1±12.42)mL；上腔静脉血流量(25.5±4.14 mL)明显低于下腔静脉血流量(42.14±9.26 mL)(t=-6.866,P<0.001),下腔静脉返流分数(9.77±1.83%)明显大于上腔静脉返流分数(6.86±1.92%)(t=4.250,P<0.001)。主、肺动脉搏出量及腔静脉回流量比值为1：1.009：0.974。4. Triple-IR结合FIESTA能清楚显示心内畸形和心脏大血管连接部畸形；3.0 TMRI测量获得的EDV、ESV、EF及FS与UCG测量值之间差异无统计学意义(P值均>0.05),二者相关性良好(r=0.727-0.99)；BGS术后,主心室EDVI、MMI及CI均较健康志愿者增大,但EF值较健康志愿者减低；电影MRI测量房室瓣返流程度与UCG测量结果呈显著正相关性(rs=0.712,P<0.001),二者对房室瓣返流严重程度测量的吻合度较好(Kappa=0.453,P=0.01)。5.BGS术后,肺、体循环血流量存在明显差异(Qs大于Qv,Qv大于Qp)；右肺血流量(56.1±10.9%)明显多于左肺血流量(43.9±10.9%)；APCF为0.89±0.47 L/min,动脉前向性血流组明显多于静脉前向性血流组；PC-MRI测量肺血流量与CE-MRI测量肺动脉发育指标呈显著正相关(r=0.456-0.698)；MRI与UCG测量腔-肺吻合口宽度、峰值流速及压差呈显著正相关(r=0.427-0.858)；心内分流量(0.61±0.29 L/min)与主心室EDVI及UCG所测房室瓣返流面积呈显著正相关(r=0.806及r=0.685)。结论：1.选择合适的Venc和编码方向是准确测量流体动力学参数的前提；3.0 TPC-MRI能客观、准确地测量流体模型的动力学信息。2.非屏气法多时相采集电影MRI能较准确地测量左心室功能,而且可用于不能配合屏气患者的心功能检查。3. PC-MRI能够准确测量肺、体循环大血管血流量,所测健康志愿者的肺、体循环血流量及返流分数对心脏病患者的检查及术后复查具有重要参考意义。4.3.0 T MRI SE序列结合电影成像能准确评价CCHD患者BGS术后主心室的结构和功能,评价房室瓣返流方面与UCG相关性良好,具有较好的临床应用价值。5.3.0 T PC-MRI测量肺血流量与CE-MRI测量肺血管发育指标具有良好相关性,二者结合可更好地评估整体肺血情况；心内分流量与EDVI及房室瓣返流关系密切；APCF对治疗决策及评估预后有重要意义。更多还原

【Abstract】 Objectives:1. To verify the effection of velocity encoding value and encoding direction in phase-contrast imaging sequence on 3.0 Tesla MR system (3.0 T PC-MRI), and to assess the accuracy and the stability for the quantitative measurement of the constant flow and variable velocity flow by using a self-made flow phantom.2. To explore the feasibility of multi-phase acquisition sequence during cardiac cycle (cine MRI) without breath-hold to evaluate the cardiac function for patients with incompatible breath-hold, and to investigate the values of left ventricular function by using this sequence.3. To measure the hemodynamic information of great vessels of pulmonary circulation and systemic circulation by using phase-contrast imaging sequence on 3.0 Tesla MR system (3.0 T PC-MRI), and to validate the accuracy of the results from PC-MRI sequence in vivo simultaneously.4. To investigate the significance of the information of the intracardiac malformations, major ventricular and atrioventricular valve function for patients with complex congenital heart disease (CCHD) underwent bidirectional Glenn shunt (BGS) from spin-echo sequence combining with cine MR imaging sequence.5. To calculate the aortopulmonary collateral flow (APCF) and intracardiac shunt flow (ICSF) for patients with bidirectional Glenn shunt (BGS) according the data from PC sequence, and to investigate the relationship of the parameters of pulmonary vessel growth and the pulmonary flows obtained from contrast-enhanced MR imaging (CE-MRI) sequence and phase-contrast imaging sequence.Materials and Methods:1. First, holding flow velocity and velocity encoding value, the dynamic information of the fluid flow was tested under different encoding directions. Second, holding flow velocity and encoding directions, the velocities of constant flow were tested under different velocity encoding values. Third, holding velocity encoding value and encoding direction, flow velocity value measured by PC-MRI sequence under different constant flow was compared with its actual value. Fourth, the curve of the variable velocity flow measured with PC-MRI sequence was compared with the curve of its actual variable velocity. T test was employed to assess the statistical significance of the difference between the results measured with PC-MRI sequence and their actual value.2. The fast imaging employing steady state acquisition (FIESTA) sequence with and without breath-hold was performed to obtain the multiphase data in 15 healthy volunteers. The left ventricular function was analyzed from the images of these volunteers on the workstation with Report Card software. The results of the function parameters of left ventricle, including end-diastolic volume (EDV), end-systolic volume (ESV), ejection fraction (EF), mean myocardial mass (MM), cardiac output (CO) and the corresponding parameters of end-diastolic volume index (EDVI), myocardial mass index (MMI) and cardiac index (CI), without breath-hold were compared with those obtained with multi-breathhold sequence. Paired-samples t-test was employed to assess their difference, and pearson’s correlation analysis was performed between the results without breath-hold sequence and with multi-breathhold sequence.3. PC-MRI sequences were performed on GE 3 tesla MR scanner in 15 healthy volunteers in order to evaluate the blood flow status in main pulmonary artery (MPA), right pulmonary artery (RPA), left pulmonary artery (LPA), ascending aorta (AA), superior vena cava (SVC) and inferior vena cava (IVC). The net flow volume and regurgitation fraction of each vessel were calculated for each volunteer during one cardiac cycle. Paired-samples t-test was employed to assess the statistical significant difference.4. Triple-IR sequence and cine MR sequence were performed in 22 patients with BGS. According to these MR images, intracardiac malformations, major ventricular and atrioventricular valve function were evaluated with Report Card software on workstation. End-diastolic volume (EDV), end-systolic volume (ESV), ejection fraction (EF), fractional shortening (FS), mean myocardial mass (MM), cardiac output (CO) and the corresponding parameters of end-diastolic volume index (EDVI), myocardial mass index (MMI) and cardiac index (CI) of major ventricular were measured with cine MR imaging sequence. The correlation analysis of the results of EDV, ESV, EF and FS obtained from ultrasound cardiography (UCG) and MRI was performed. The results of EDVI, MMI, CI and EF of BGS in the patients and healthy volunteers were compared and independent-samples t-test was employed to evaluate the statistical significance of the results’ difference. Spearman rank correlation coefficients were employed to analyze the difference of the regurgitation degree of atrioventricular valve obtained with cine MRI and UCG.5. PC-MRI sequence was performed to measure the flow of great vessels of right pulmonary artery (RPA), left pulmonary artery (LPA), ascending aorta (AA), superior vena cava (SVC) and inferior vena cava (IVC), and the quantity of AA (Qs) per minute (Qs), the quantity of pulmonary per minute (Qp), and the quantity of venous return per minute (Qv) were calculated by using Report Card software. APCF was calculated as the difference of Qs and Qp, and ICSF were calculated as the difference of Qv and Qp. CE-MR imaging sequence was performed to assess the pulmonary vessels growth. The relationship of parameters of pulmonary vessels growth and corresponding blood flow was evaluated with the correlation analysis. The statistical significant difference of pulmonary blood flow and APCF at the different antegrade flow groups was evaluated with independent samples t-test. The results of the width of superior cavo-pulmonary anastomosis measured with MRI and UCG were evaluated with paired-samples t-test and pearson’s correlation analysis. The relationship of ICSF and EDVI obtained with MRI was evaluated with correlation and regression analysis, and so did the relationship of ICSF and the regurgitation area of atrioventricular valve obtained with UCG.Results:1. Only on the velocity encoding direction of SLICE or SI (superior and inferior) could the direction of proton flowing be correctly detected. There was no significant difference between the measured velocities with PC-MRI sequence under different flow rates and the corresponding actual flow velocities (t=-0.861, P=0.405) while the velocity encoding value was appropriately greater than the actual flow velocity. A significantly positive correlation (r=0.999, P＜0.001) was demonstrated between the measured velocities and the actual velocities. There was good consistency among the measured maximum, minimum, average velocities and the corresponding actual variable flow velocities.2. The EDV, ESV, EF, MM, EDVI, MMI, CO and CI measured with and without breath-hold multi-phase acquisition sequence were (121.97±17.53) ml and (123.85±19.48) ml, (45.57±9.18) ml and (46.64±8.34) ml, (62.39±4.67)% and (62.65±5.12)%, (106.25±18.07) g and (105.63±19.20) g, (69.58±6.53) ml/m2 and (70.44±7.11) ml/m2, (60.17±6.34) g/m2 and (60.83±5.45) g/m2, (5.69±0.88) L/min and (5.70±0.78) L/min, (3.27±0.30) L/min/m2 and (3.26±0.24) L/min/m2, respectively. No statistical difference was found between the results of the two methods (all P>0.05), but good correlationship between the results of the two methods was demonstrated (r=0.633-0.957, all P< 0.01).3. The flow volume of MPA, RPA and LPA during a cardiac cycle was (69.42±12.35) ml, (38.78±6.81) ml and (32.27±6.49) ml, respectively. The flow volume of RPA was significantly higher than that of LPA (t=3.092, P= 0.004). The percentage of flow volume of RPA and LPA was 55.4% and 44.6%, respectively. The regurgitation fraction of LPA was significantly higher than that of RPA (t=5.502, P=0.001). The average flow volume of AA in one cardiac cycle was (70.1±12.42) ml. The average flow volume of SVC (25.5±4.14 ml) was significantly lower than that of IVC (42.14±9.26 ml) (t=-6.866, P＜0.001). The regurgitation fraction of IVC (9.77±1.83%) was significantly higher than that of SVC (6.86±1.92%) (t=4.250, P<0.001). The ratio of flow volumes of AA, MPA and vena cava were 1:1.009:0.974.4. All intracardiac malformations and cardiovascular conjunction were demonstrated clearly on the images of triple-IR sequence with cine MR sequence. No statistical difference was found between the results of EDV, ESV, EF, FS obtained with cine MR imaging sequence and those with UCG (all P>0.05). A good correlation (r=0.727-0.99, all P< 0.01) was found between the two methods. The value of EDVI, MMI and CI of the patients with BGS was significantly higher than that of healthy volunteers, but the value of EF was lower. The positive correlation of the regurgitation degree of atrioventricular valve between two results obtained with MRI and UCG was found significantly (rs=0.712, P＜0.001).5. The difference of the flow volume of systemic circulation and pulmonary circulation was found significantly. We found that Qs was higher than Qv, and Qv was higher than Qp. The flow volume of RPA (56.1±10.9%) was significantly higher than that of LPA (43.9±10.9%). The flow volume of APCF ranged from 0.21-1.53 L/min (mean 0.89 L/min), and the flow volume of antegrade venous blood flow group was significantly higher than that of antegrade arterial blood flow group. The positive relationship of the flow volume and the growth parameters of pulmonary arteries was found (r=0.456-0.698). The results of width, peak flow velocity and gradient pressure of superior cavopulmonary anastomosis obtained by MRI were closely correlated with those of UCG (r=0.427-0.858, all P< 0.05). The positive relationship between ICSF and EDVI obtained by cine MR imaging sequence, and the positive relationship between ICSF and the regurgitation area of atrioventricular valve obtained by UCG were found (r=0.685, and r=0.806).Conclusions:1. PC-MRI sequence on 3.0 T MR system can evaluate the parameters of the flow of the flow phantom accurately when the velocity encoding value and the encoding direction are selected correctly.2. Left ventricular function in healthy people and the patients with incompatibile breath-hold can be evaluated with multi-phase acquisition without breath-hold sequence correctly.3. The flow volume of great vessels of pulmonary circulation and systemic circulation can be accurately measured with PC-MRI sequence. The flow volume and regurgitation fraction of great vessels in healthy volunteers may be used as a reference.4. The cardiac malformation and function of the BGS patients could be accurately evaluated with cine MR imaging sequence and spin-echo sequence. A good correlationship was demonstrated in evaluating the degree of regurgitation of atrioventricular valve between cine MR imaging sequence and UCG.5. The SPCF and ICSF in patients with BDG can be reliably measured with PC-MRI sequence on 3.0 tesla MR system. The good correlation of the flow volume obtained by PC-MRI sequence and the growth parameters obtained by CE-MRI for pulmonary vessels are demonstrated, and so does the ICSF with EDVI and with the regurgitation area of atrioventricular valve obtained by UCG.更多还原