Dixon与SPAIR技术在颈部MRI中的脂肪抑制效果对比研究

Comparison of fat suppression effects between Dixon and SPAIR techniques in the neck MRI

目的探讨水脂分离技术(Dixon)与精准频率反转恢复技术(SPAIR)颈部脂肪抑制效果。方法前瞻性收集2018年6月至12月南京医科大学附属逸夫医院行颈椎MRI检查的35例患者,男16例,女19例,年龄23~75(46±13)岁,行Dixon和SPAIR脂肪抑制扫描。扫描序列包括Dixon容积内插体部检查(VIBE)-T1WI、SPAIR VIBE-T1WI、Dixon快速自旋回波(TSE)-T2WI和SPAIR TSE-T2WI。主观评价将脂肪抑制质量分为0~4级,将甲状腺组织显示清晰的分别计数为甲Dixon和甲SPAIR。采用Fisher精确检验主观评价,采用配对t检验或秩和检验比较2种方法在下颌角、声门、甲状腺及胸锁关节层面的噪声、信噪比(SNR)和对比噪声比(CNR)。结果主观评价:Dixon VIBE-T1WI中0级29例,1级4例,2级2例,SPAIR VIBE-T1WI组中3级4例,4级31例,差异有统计学意义(P<0.05);Dixon TSE-T2WI中0级28例、1级5例、2级2例,无3和4级,SPAIR TSE-T2WI中3级2例、4级33例,无1和2级,差异有统计学意义(P<0.05)。VIBE-T1WI中甲Dixon35例,甲SPAIR4例,差异有统计学意义(P<0.05);TSE-T2WI中甲Dixon35例,甲SPAIR3例,差异有统计学意义(P<0.05)。客观评价:下颌角层面,VIBE-T1WI序列,Dixon图像胸锁乳突肌SNR、头夹肌SNR和CNR大于SPAIR图像,差异有统计学意义(P<0.05);TSE-T2WI序列,Dixon图像噪声、胸锁乳突肌SNR和头夹肌SNR大于SPAIR图像,差异有统计学意义(P<0.05)。声门层面,VIBE-T1WI序列,Dixon图像中胸锁乳突肌SNR、肩胛提肌SNR大于SPAIR图像,差异有统计学意义(P<0.05);TSE-T2WI序列,Dixon图像中噪声和胸锁乳突肌SNR大于SPAIR图像,差异有统计学意义(P<0.05)。甲状腺层面,VIBE-T1WI序列,Dixon图像噪声小于SPAIR图像,Dixon图像胸锁乳突肌SNR和肩胛提肌SNR大于SPAIR图像,差异具有统计学意义(P<0.05);TSE-T2WI序列,Dixon图像中胸锁乳突肌SNR大于SPAIR图像,差异有统计学意义(P<0.05)。胸锁关节层面,VIBE-T1WI组和TSE-T2WI组,Dixon图像与SPAIR图像噪声、SNR和CNR差异均无统计学意义(P>0.05)。结论颈部MRI扫描中,Dixon较SPAIR能获得更好的图像质量和脂肪抑制效果。

Objective To investigate the effect of water-fat separation(Dixon)and spectrally selective attenuated inversion recovery(SPAIR)on fat suppression in the neck.Methods From June to December 2018,35 patients,including 16 males and 19 females,aged 23-75(46±13)years old underwent MRI examinations of cervical vertebrae in Sir Run Run Hospital Affiliated to Nanjing Medical University.The imaging data of Dixon and SPAIR fat suppression scanning were prospectively collected.The scanning sequences included Dixon VIBE-T1WI,SPAIR VIBE-T1WI,Dixon TSE-T2WI and SPAIR TSE-T2WI.According to the subjective evaluation,the quality of fat suppression was divided into 0-4 grades.The cases with clear depiction of the thyroid tissue on Dixon or SPAIR were counted.Fisher exact test and paired t test or rank sum test were used to compare the noise,SNR and CNR of mandible angle,glottis,thyroid and sternoclavicular joint.Results There were 29 cases of grade 0,4 cases of grade 1,2 cases of grade 2 on Dixon VIBE-T1WI,4 cases of grade 3 and 31 cases of grade 4 on SPAIR VIBE-T1WI group,and the difference was statistically significant(P<0.05);28 cases of grade 0,5 cases of grade 1 and 2 cases of grade 2 on Dixon TSE-T2WI,33 cases of grade 3 and 4 on SPAIR TSE-T2WI,and the difference was statistically significant(P<0.05).There were 35 cases with clear depiction of the thyroid tissues on DIXON and 4 cases on SPAIR VIBE-T1WI(P<0.05);there were 35 cases on DIXON and 3 cases on SPAIR TSE-T2WI(P<0.05).At the level of mandible angle,for the SNR of sternocleidomastoid muscle,SNR and CNR of splenius capitis muscle on VIBE-T1WI,Dixon images were better than those on SPAIR images,and the difference was statistically significant(P<0.05);the noise,SNR of sternocleidomastoid and splenius capitis muscle on TSE-T2WI images were more than those in SPAIR images,and the difference was statistically significant(P<0.05).At the level of glottis,the SNR of sternocleidomastoid muscle and levator scapulae muscle in the sequence of VIBE-T1WI and Dixon images were significantly higher than those on the images of SPAIR(P<0.05);the noise and SNR of sternocleidomastoid muscle on Dixon TSE-T2WI images were significantly higher than those on the images of SPAIR(P<0.05).At the thyroid level,the noise of Dixon VIBE-T1WI image was less than that of SPAIR image,the SNR of sternocleidomastoid muscle and levator scapulae muscle on Dixon images was greater than that of SPAIR image,and the difference was statistically significant(P<0.05);For TSE-T2WI sequence,the SNR of sternocleidomastoid muscle on Dixon image was greater than that on SPAIR image,and the difference was statistically significant(P<0.05).At the level of sternoclavicular joint,there were no significant differences for noise,SNR and CNR between the two groups(P>0.05).Conclusion Dixon outperforms SPAIR in obtaining better image quality and suppression effect in terms of neck MRI scanning.