Shea Steven M, Fieno David S, Schirf Brian E, Bi Xiaoming, Huang Jie, Omary Reed A, Li Debiao
Department of Biomedical Engineering, Radiology, and Preventive Medicine, Northwestern University, Evanston, Ill 60611, USA.
Radiology. 2005 Aug;236(2):503-9. doi: 10.1148/radiol.2362040149.
To assess the ability of a T2-prepared steady-state free precession blood oxygen level-dependent (BOLD) magnetic resonance (MR) imaging sequence to depict changes in myocardial perfusion during stress testing in a dog stenosis model.
Study was approved by the institutional Animal Care and Use Committee. A hydraulic occluder was placed in the left circumflex coronary artery (LCX) in 10 dogs. Adenosine was administered intravenously to increase coronary blood flow, and stenosis was achieved in the LCX with the occluder. A T2-prepared two-dimensional steady-state free precession sequence was used for BOLD imaging at a spatial resolution of 1.5 x 1.2 x 5.0 mm3, and first-pass perfusion images were acquired for visual comparison. Microspheres were injected to provide regional perfusion information. Mixed-effect regression analysis was performed to assess normalized MR signal intensity ratios and microsphere-measured perfusion differences. For the same data, 95% prediction intervals were calculated to determine the smallest perfusion change detectable. Means +/- standard deviations were calculated for myocardial regional comparison data. A two-tailed Student t test was used to determine if significant differences (P < .01) existed between different myocardial regions.
Under maximal adenosine stress, MR clearly depicted stenotic regions and showed regional signal differences between the left anterior descending coronary artery (LAD)-fed myocardium and the stenosed LCX-fed myocardium. Visual comparisons with first-pass images were also excellent. Regional MR signal intensity differences between LAD and LCX-fed myocardium (1.24 +/- 0.08) were significantly different (P < .01) from differences between LAD and septal-fed myocardium (1.02 +/- 0.07), which was in agreement with microsphere-measured flow differences (LAD/LCX, 3.38 +/- 0.83; LAD/septal, 1.26 +/- 0.49). The linear mixed-effect regression model showed good correlation (R = 0.79) between MR differences and microsphere-measured flow differences.
On T2-prepared steady-state free precession BOLD MR images in dogs, signal intensity differences were linearly related to flow differences in myocardium, with a high degree of correlation.
radiology.rsnajnls.org/cgi/content/full/236/2/503/DC1
评估一种T2准备的稳态自由进动血氧水平依赖(BOLD)磁共振(MR)成像序列在犬狭窄模型应激试验期间描绘心肌灌注变化的能力。
本研究经机构动物护理和使用委员会批准。在10只犬的左旋冠状动脉(LCX)中放置液压闭塞器。静脉注射腺苷以增加冠状动脉血流,并用闭塞器使LCX出现狭窄。使用T2准备的二维稳态自由进动序列进行BOLD成像,空间分辨率为1.5×1.2×5.0 mm3,并采集首过灌注图像用于视觉比较。注射微球以提供区域灌注信息。进行混合效应回归分析以评估标准化MR信号强度比和微球测量的灌注差异。对于相同数据,计算95%预测区间以确定可检测到的最小灌注变化。计算心肌区域比较数据的均值±标准差。使用双尾Student t检验确定不同心肌区域之间是否存在显著差异(P < 0.01)。
在最大腺苷应激下,MR清晰描绘了狭窄区域,并显示了左前降支冠状动脉(LAD)供血的心肌与狭窄的LCX供血的心肌之间的区域信号差异。与首过图像的视觉比较也非常好。LAD和LCX供血的心肌之间的区域MR信号强度差异(1.24±0.08)与LAD和间隔供血的心肌之间的差异(1.02±0.07)有显著差异(P < 0.01),这与微球测量的血流差异一致(LAD/LCX,3.38±0.8;LAD/间隔,1.26±0.49)。线性混合效应回归模型显示MR差异与微球测量的血流差异之间具有良好的相关性(R = 0.79)。
在犬的T2准备的稳态自由进动BOLD MR图像上,信号强度差异与心肌血流差异呈线性相关,且相关性较高。
radiology.rsnajnls.org/cgi/content/full/236/2/503/DC1