Song Kwang-Sup, Jang Eui-Chan, Jung Ho-Joong, Kim Kyung-Woon, Yu Hyeon
Department of Orthopedic Surgery, Chung-Ang University, College of Medicine, Seoul, Korea.
J Spinal Disord Tech. 2008 Dec;21(8):569-74. doi: 10.1097/BSD.0b013e3181639b3b.
Retrospective study.
To determine whether magnetic resonance myelography (MRM) improves the evaluation of the severity of stenosis in patients with multilevel lumbar stenosis.
MRM shows a similar image as myelography in a noninvasive manner.
One hundred patients over 50 years of age with multiple lumbar stenosis who were prospectively referred for MR imaging (MRI) with MRM were enrolled in the study. The most severe stenotic segment and the degree of stenosis of that segment, as assessed by the extent of remaining subarachnoidal space (1: normal to 50%; 2: over 50% but not a total blockage; 3: total blockage) were evaluated in a blinded manner by 2 observers. Conventional MRI (class A), MRM (class B), and MRI combined with MRM (class C) were evaluated independently and interobserver and intraobserver reliability were assessed.
In the selection of the most severe segment and degree of stenosis, for both observers, the consensus between class (B) and class (C), was higher than that of class (A) and class (C). The average kappa values for interobserver agreement in the selection of the most severe segment/assessing the degree of stenosis for classes (A), (B), and (C) were 0.649/0.727, 0.782/0.771, and 0.832/0.784, respectively. Intraobserver kappa values were also highest for class (B), followed by class (C), and then class (A). Observations were within the range of "almost perfect" (0.81< or =kappa< or =1), with the exception of the selection of the most severe segment in class (A) by one of the observers.
When employed in routine practice, MRM could be of value for improving observer reliability in the assessment of severity of stenosis in multiple lumbar stenosis.
回顾性研究。
确定磁共振脊髓造影（MRM）是否能改善对多节段腰椎管狭窄症患者狭窄严重程度的评估。
MRM以非侵入性方式显示与脊髓造影相似的图像。
100例年龄超过50岁的多节段腰椎管狭窄患者被前瞻性纳入本研究，这些患者接受了带有MRM的磁共振成像（MRI）检查。由两名观察者以盲法评估最严重狭窄节段及其狭窄程度，根据蛛网膜下腔剩余范围进行评估（1：正常至50%；2：超过50%但未完全阻塞；3：完全阻塞）。对传统MRI（A类）、MRM（B类）和MRI联合MRM（C类）进行独立评估，并评估观察者间和观察者内的可靠性。
在选择最严重节段和狭窄程度方面，对于两名观察者而言，B类和C类之间的一致性高于A类和C类之间的一致性。A类、B类和C类在选择最严重节段/评估狭窄程度时观察者间一致性的平均kappa值分别为0.649/0.727、0.782/0.771和0.832/0.784。观察者内kappa值在B类中也最高，其次是C类，然后是A类。观察结果在“几乎完美”（0.81≤kappa≤1）范围内，但有一名观察者对A类中最严重节段的选择除外。
在常规实践中应用时，MRM可能有助于提高观察者对多节段腰椎管狭窄症狭窄严重程度评估的可靠性。
