Purpose

Studies comparing back pain patients and controls on continuous intervertebral kinematics have shown differences using univariate parameters. Hitherto, multivariate approaches have not been applied to this high dimensional data, risking clinically relevant features being undetected. A multivariate re-analysis was carried out to estimate main modes of variation, and explore group differences.

Methods

40 participants with mechanical back pain and 40 matched controls underwent passive recumbent quantitative videofluoroscopy. Intervertebral angles of L2/3 to L4/5 were obtained for right and left side-bending, extension, and flexion. Principal components analysis (PCA) was used to identify the main modes of variation, and to obtain a lower dimensional representation for comparing groups. Linear discriminant analysis (LDA) was used to identify how groups differed.

Results

PCA identified three main modes of variation, all relating to range of motion (ROM) and its distribution between joints. Significant differences were found for coronal plane motions only (right: p = 0.02, left: p = 0.03) . LDA identified a shift in ROM to more cranial joints in the back pain group.

Conclusion

The results confirm altered motion sharing between intervertebral joints in back pain, and provides more details about this. Further work is required to establish how these findings lead to pain, and so strengthen the theoretical basis for treatment and management of this condition.

中文翻译：

---

慢性中下腰痛的被动椎间运动特征：多变量分析。

目的

在连续椎间运动学上比较背痛患者和对照组的研究表明使用单变量参数存在差异。迄今为止，尚未对这种高维数据应用多变量方法，这可能会导致临床相关特征未被检测到。进行了多变量重新分析以估计主要变异模式，并探索组间差异。

方法

40 名患有机械性背痛的参与者和 40 名匹配的对照组接受了被动卧位定量视频透视。获得左右侧弯曲、伸展和屈曲的椎间角为 L2/3 至 L4/5。主成分分析 (PCA) 用于识别主要变异模式，并获得用于比较组的较低维度表示。使用线性判别分析 (LDA) 来确定组之间的差异。

结果

PCA 确定了三种主要的变化模式，都与运动范围 (ROM) 及其在关节之间的分布有关。仅在冠状平面运动中发现显着差异（右：p  = 0.02，左：p  = 0.03）。LDA 发现背痛组 ROM 向更多颅关节转移。

结论

结果证实了背痛中椎间关节之间运动共享的改变，并提供了更多关于此的细节。需要进一步的工作来确定这些发现如何导致疼痛，从而加强治疗和管理这种疾病的理论基础。