术前MRI测量脊髓受压程度与脊髓型颈椎病手术疗效的相关性

The correlation between preoperative compression degree of the spinal cord assessed by magnetic resonance imaging and the results of surgery for cervical spondylotic myelopathy

目的：探讨术前应用MRI测量的脊髓受压程度与脊髓型颈椎病手术效果的相关性。方法：2006年2月-2010年7月在北京大学第三医院骨科行颈前路或后路减压手术的115例脊髓型颈椎病患者被纳入研究,其中男70例,女45例,年龄24-85岁,平均56岁。随访时间24-60个月,平均29个月。在术前MRI矢状位T2像上观察脊髓受压节段数,并在受压最重节段测量并计算硬膜囊中矢径/椎体中矢径（中矢径比值）;在相同节段MRI轴位T2像上测量并计算脊髓矢状径/硬膜囊矢状径×100%（矢状径占有率）、脊髓水平径/硬膜囊水平径×100%（水平径占有率）、脊髓横截面积/硬膜囊横截面积×100%（硬膜囊占有率）。应用改良JOA（mJOA）评分评价患者脊髓功能,应用评分改善率评价手术效果。将患者脊髓受压节段数、中矢径比值、矢状径占有率、水平径占有率、硬膜囊占有率及术前mJOA评分与术后24个月mJOA评分及评分改善率进行相关性分析,并以术后24个月mJOA评分改善率为因变量,以上述各MRI测量指标及术前mJOA评分为自变量,向后线性回归分析,得到拟合mJOA评分及评分改善率,并采用Pearson相关系数检验实际mJOA评分改善率与拟合mJOA评分改善率的相关性。结果：脊髓受压1个节段37例,2个节段17例,3个节段15例,4个节段25例,5个节段21例。MRI中矢径比值为0.426±0.097,矢状径占有率为（79.1±8.4）%,水平径占有率为（76.2±7.3）%,硬膜囊占有率为（54.6±16.2）%。mJOA评分由术前的12.1±2.9分增加至术后的14.7±2.0分（P〈0.001）;mJOA评分改善率为（43.0±55.7）%（-200%-100%）。术后24个月mJOA评分及评分改善率与术前mJOA评分、中矢径比值、矢状径占有率均显著相关（P〈0.05）,与受压节段数、水平径占有率及硬膜囊占有率无显著相关性（P〉0.05）。对术后24个月mJOA评分向后线性回归得到方程：术后24个月mJOA评分=4.202＋0.346×术前mJOA评分＋4.973×中矢径比值＋0.053×矢状径占有率（%）。对mJOA评分改善率向后线性回归得到方程：mJOA评分改善率（%）=-30.348＋115.875×中矢径比值＋1.226×矢状径占有率（%）-5.993×术前mJOA评分,依据上述回归方程计算的拟合mJOA评分改善率与实际mJOA评分改善率显著相关（R^2=0.138,P〈0.001）。结论：术前应用MRI测量的脊髓在矢状径方向的受压程度和脊髓型颈椎病的手术效果显著相关。

Objectives： To investigate the correlation between preoperative compression degree of the spinal cord assessed by magnetic resonance imaging and the results of surgery for cervical spondylotic myelopathy.Methods： From February 2006 to July 2010, 115 patients with cervical spondylotic myelopathy in Peking University Third Hospital undergoing anterior or posterior decompression were included in this study. There were 75 males and 40 females with the age ranged from 24 to 85 years（average, 56 years）. All patients were followed up for at least 24 months（average, 29 months）. The number of compression levels was measured on T2-weighted sagittal MRI, then the ratio of the sagittal diameter of the spinal cord to the sagittal diameter of the dural sac was calculated as MRI Pavlov ratio at the site of maximal cord compression. The transversearea, transverse diameter and sagittal diameter of the spinal cord and the dural sac at the site of maximal cord compression were measured on T2-weighted axial MRI. Then the occupation ratio was calculated as the ratio of the area of the spinal cord to the area of the dural sac, the sagittal/transverse occupation ratio was calculated as the ratio of the sagittal/transverse diameter of the spinal cord to the sagittal/transverse diameter of the dural sac as percentage. The severity of myelopathy was evaluated by the modified Japanese Orthopaedic Association（mJOA） score, and recovery was determined by recovery rate（RR）. Imaging fidings and preoperative mJOA（pre-mJOA） score were then correlated with 24-month postoperative mJOA（post-mJOA） score and RR. Then a linear regression analysis was established regarding the 24-month post-mJOA score and RR as dependent variables, and the predictive RR was correlated with the actual RR. Results： The spinal cord was compressed at 1 level in 37 cases, 2 levels in 17 cases, 3 levels in 15 cases, 4 levels in 25 cases and5 levels in 21 cases. The MRI Pavlov ratio was 0.426±0.097, the sagittal and the transverse occupation ratio was（79.1±8.4）% and（76.2 ±7.3）%, and the occupation ratio was（54.6 ±16.2）%. The mJOA score increased from the preoperative 12.1±2.9 to the postoperative 14.7±2.0（P〈0.001）, and the RR was（43.0±55.7）%[（-200%）-100%]. The post-mJOA score and the RR correlated significantly with the pre-mJOA score, the MRI Pavlov ratio and the sagittal occupation ratio（P〈0.05）, but not correlated with the occupation ratio or the transverse occupation ratio significantly（P〈0.05）. The multiple regression equations to predict the results of surgery were as follows： post-mJOA=4.202＋0.346×pre-mJOA＋4.973×MRI Pavlov ratio＋0.053×sagittal occupation ratio（%）;RR=-30.348＋ 115.875×MRI Pavlov ratio＋1.226×sagittal occupation ratio（%）-5.993×pre-mJOA. The predictive and actual RRs were correlated significantly（R^2=0.138, P〈0.001）. Conclusions： The compression degree of the spinal cord at the sagittal plane assessed preoperatively by MRI correlates significantly with the surgical effects of cervical spondylotic myelopathy.