心肌磁共振多b值DWI的初步探讨

A preliminary study of magnetic resonance myocardial multi-b values diffusion weighted imaging

目的:探讨心肌磁共振多b值扩散加权成像的可行性。方法:11例健康志愿者行心脏多b值扩散加权扫描,由两位放射科医师对图像质量进行评估。符合要求的图像在标准横断位和心脏短轴位上分别测量室间隔前部、后部、上部、下部及游离壁的ADCslow值、ADCfast值和f值。所得数据进行Bland-Altman分析和Pearson相关分析。结果:8例(73%)志愿者检查成功,共获得了16组符合定量测量要求的图像,其中Ⅰ级图像6组(6/16,37.5%),Ⅱ级图像10组(10/16,62.5%),在标准横轴面和心脏短轴位上分别测量室间隔前部、后部、上部、下部及游离壁的ADCslow值(0.0027±0.0009)×10-3～(0.0051±0.0029)×10-3 mm2/s、ADCfast值(0.1478±0.0832)×10-3～(0.1935±0.0817)×10-3 mm2/s和f值(0.1377±0.1380)～(0.2957±0.1166),根据Bland-Altman图和Pearson相关分析得到参数测量在观察者间具有良好的一致性[95%一致性界限(-1.5599,1.4510),相关系数r=0.814]。结论:在呼吸训练和心率控制的前提下,经过恰当的参数选择,能获得清晰的心肌多b值扩散加权图像,且参数测量在观察者间具有良好的一致性。

Objective：To investigate the feasibility of myocardial magnetic resonance multi-b values diffusion-weighted imaging. Methods：Eleven healthy volunteers underwent myocardial multi-b values diffusion-weighted imaging. Image quality was evaluated by two radiologists independently. ADCslow, ADCfast, and fraction of ADCfast, were measured in anterior and posterior ventricular septal and free wall on the standard axis and cardiac short axis images. The agreement of two radiologists was evaluated by Bland-Altman plot and Pearson correlation analysis. Results：8 volunteers （73%） obtained images of Class Ⅰ （6/16,37.5%）or Class Ⅱ （10/16,62.5%） on standard axis and cardiac short axis images. According to the Bland-Airman plot and Pearson correlation analysis, ADCslow value [（0. 0027 ± 0. 0009） × 10^-3 - （0. 0051 ± 0. 0029 ） × 10 ^-3 mm^2/s ], ADCfast value [（0. 1478±0. 0832） × 10 ^-3 -（0. 1935±0. 0817） × 10^-3mm^2/s] and fraction of ADCfast[（0. 1377±0. 1380）- （0. 2957 ±0. 1166）] measurement had a good inter-observer agreement [95 % limits of agreement （-1. 5599,1. 4510）, correlation coefficient r= 0. 814]. Conclusion： With breathe training, heart rate control and appropriate scanning parameters application,it is feasible to obtain myocardial multi- b values diffusion-weighted images,and have a good inter-observer agreement of the measurements.