Abstract: Objective The effects of different contrast agent injection rate on the quality of 3D enhanced MRA images were analyzed, and the optimal contrast agent injection rate on high-quality cervical enhanced MRA images was discussed. Methods Patients with suspected cervical vascular lesions who underwent enhanced cervical MRA were randomly divided into 3 groups according to the injection rate. There were 22 cases in group A（1 mL/s）, 25 cases in group B（2 mL/s）, and 20 cases in group C（3 mL/s）. The quality of arterial images and the degree of venous contamination from raw data were graded by two attending physicians, and the mean scores of the whole artery and venous contamination were calculated. The brachiocephalic trunk, left common carotid artery, left subclavian artery, 1 cm below the bifurcation of bilateral common carotid artery and bilateral same-level vertebral artery, bilateral internal carotid cavernous sinus segment （C4） and basilar artery signal intensity （SI） were measured by another attending physician. Results The average arterial signal of groups A, B and C were 584, 720 and 690 respectively. The mean value of arterial contrast noise ratio （CNR） are 222.50, 272.94 and 280.22 respectively. The above three groups are statistically significant （SI H=8.29, P=0.016, CNR H=7.56, P=0.023, P<0.05）. There was statistical difference in arterial, venous and overall scores （P<0.05）. There was statistical difference not only between groups A and B, but also group A and C in arterial scoring （P<0.017）, but there was no statistical difference between groups B and C （P>0.017）. There was statistical difference between groups A and C in venous contamination score. There was no statistical difference between Group A and Group B as well as that Group B and Group C （P>0.017）, there was no statistical difference between Group A and Group B and Group C （P<0.017）, and there was no statistical difference between Group B and Group C （P>0.017）. Conclusion With the increase of injection rate, the signal intensity of carotid artery increased, the image signal-to-noise ratio increased, but the venous contamination did not increase. Therefore, injection rate of 3 mL/s was recommended for cervical enhanced MRA.

Key words: cervical enhanced MRA, injection rate, venous contamination, image quality

摘要： 目的 分析不同对比剂注射流率对颈部动脉增强血管成像（3D CE-MRA）图像质量的影响,探讨获得高质量颈部CE-MRA图像的最优对比剂注射流率。方法 收集怀疑颈部血管病变而行颈部增强MRA扫描的患者,根据注射流率不同分为三组,其中A组1 mL/s共22例,B组2 mL/s共25例,C组3 mL/s共20例。由两名主治医师对各节段的动脉图像质量及静脉污染程度分别进行评分,计算整体动脉平均分和静脉污染平均分;分别测量头臂干、左颈总动脉、左锁骨下动脉、双侧颈总动脉分叉下1 cm处及双侧同水平椎动脉、双侧颈内动脉海绵窦段（C4）及基底动脉的信号强度（SI）,计算整体动脉血管信号平均值及整体血管、对比噪声比（CNR）。结果 A、B、C三组动脉血管信号平均值分别为584、720、690;动脉血管对比噪声比（CNR）平均值分别为222.50、272.94、280.22,以上指标三组总体具有统计学意义（SI H=8.29,P=0.016;CNR H=7.56,P=0.023）。动脉、静脉及整体评分均具有统计学差异（P<0.05）,其中动脉评分A组与B组及A组与C组具有统计学差异（P<0.017）,B组与C组不具有统计学差异（P>0.017）,静脉污染评分A组与C组具有统计学差异（P<0.017）,而A组与B组及B组与C组不具有统计学差异（P>0.017）,整体评分中A组与B组及A组与C组具有统计学差异（P<0.017）,B组与C组不具有统计学差异（P>0.017）。结论 随注射流率的增加颈动脉血管的信号强度增加,图像信噪比增加,而静脉污染没有增加,故行颈部CE-MRA时推荐注射流率为3 mL/s可取得高质量的血管成像效果。

关键词: 颈部增强血管成像, 注射流率, 静脉污染, 图像质量