腰骶丛神经根成像中T23D SPACE序列的技术优化研究

Study on Optimization of T23D SPACE Sequence in Lumbosacral Plexus Nerve Root Imaging

目的:探讨T23D SPACE序列在腰骶丛神经根中的应用及其参数优化。方法:前瞻性招募并收集46名正常志愿者的临床与影像资料,所有志愿者均行常规T23D SPACE(方法A)、优化后T23D SPACE(方法B)磁共振扫描,并比较2种方法的扫描时间,第4~5腰神经、第1~4骶神经、坐骨神经的神经显示情况,第5腰神经根节中、节前、节后的信噪比(SNR)、对比度噪声比(CNR)_(神经-肌肉)、对比度(CR)_(神经-肌肉值)。优化后T23D SPACE序列(方法B)主要优化了TR、TE、加速因子(回波链)、脂肪抑制方式、血液抑制方式、层厚等参数。结果:(1)方法A扫描时间为356 s,方法B扫描时间为229 s。(2)在神经显示评分比较中,第4、5腰神经,第1~4骶神经和坐骨神经的显示评分在2种方法之间差异无统计学意义。(3)第5腰神经根的节中SNR值(286.842±75.822 vs 376.784±111.880)、CNR_(神经-肌肉值)(389.199±106.824 vs 522.683±159.883)、CR_(神经-肌肉值)(0.798±0.037 vs 0.830±0.038)在2种方法间比较,差异均有统计学意义(P<0.05)。(4)第5腰神经根的节前SNR值(198.758±52.966 vs 260.378±79.631)、CNR_(神经-肌肉值)(254.720±74.904 vs344.948±112.041)、CR_(神经-肌肉值)(0.718±0.070 vs 0.762±0.056)在2种方法间比较,差异均有统计学意义(P<0.05)。(5)第5腰神经根的节后SNR值(161.400±46.883 vs 206.849±59.706)、CNR_(神经-肌肉值)(197.684±63.776 vs 263.240±80.910)、CR_(神经-肌肉值)(0.663±0.068 vs 0.711±0.058)在2种方法间比较,差异均有统计学意义(P<0.05)。结论:常规T23D SPACE序列与优化后的序列均可以显示腰骶神经,优化后扫描更快,神经显示中第5腰神经根的信噪比、对比度均比常规T23D SPACE序列高。

Purpose:To explore the application and parameter optimization of T23D SPACE sequence in displaying the lumbosacral plexus nerve root.Methods:The clinical and imaging data of 46 normal volunteers were prospectively recruited and collected.All volunteers underwent conventional T23D SPACE(method A)and optimized T23D SPACE(method B)magnetic resonance scanning.The scanning time,nerve display of the 4th-5th lumbar nerve,the 1st-4th sacral nerve and the sciatic nerve,and values of signal-to-noise ratio(SNR),contrast-tonoise ratio(CNR)_(nerve-muscle) and contrast ratio(CR)_(nerve-muscle) in the middle,preganglionic,postganglionic and segments of the 5th lumbar nerve root between the two methods were compared.The parameters of optimized T23D SPACE(method B),such as TR,TE,turbo factor(echo chain),fat suppression mode,blood suppression mode,and slice thickness were optimized.Results: (1) The scanning time of method A was 356 s and the scanning time of method Bwas 229 s. (2) No statistical difference was found in the display scores of the 4th-5th lumbar nerve, the 1st-4th sacralnerve and the sciatic nerve between the two methods. (3) There were significant differences in the middle SNR(286.842±75.822 vs 376.784±111.880), CNR_(nerve-muscle) (389.199±106.824 vs 522.683±159.883) and CR_(nerve-muscle) (0.798±0.037 vs 0.830±0.038) value of the 5th lumbar nerve root between the two methods (P<0.05). (4) There weresignificant differences in the preganglionic SNR (198.758±52.966 vs 260.378±79.631), CNR_(nerve-muscle) (254.720±74.904vs 344.948±112.041) and CR_(nerve-muscle) (0.718±0.070 vs 0.762±0.056) value of the 5th lumbar nerve root between thetwo methods (P<0.05). (5) There were significant differences in the postganglionic SNR (161.400±46.883 vs206.849±59.706), CNR_(nerve-muscle) (197.684±63.776 vs 263.240±80.910) and CR_(nerve-muscle) (0.663±0.068 vs 0.711±0.058)value of the 5th lumbar nerve root between the two methods (P<0.05). Conclusion: Both the conventional T2 3DSPACE sequence and the optimized sequence can display the lumbosacral nerve. After optimization, the scanning isfaster. The SNR and contrast of the fifth lumbar nerve root in the nerve display are higher than that of the conventionalT2 3D SPACE sequence.