Abstract:

Objective　To construct and validate a risk model for adjacent segment disease (ASD) after lumbar fusion surgery by analyzing measurements of the spinal pelvic sagittal plane. Methods　A retrospective analysis was conducted on the data of 197 patients who underwent transforaminal lumbar interbody fusion at the First Affiliated Hospital of Xi'an Jiaotong University Yulin Hospital from October 2020 to October 2023. After one year of follow-up, the patients finally included in the study were divided into an ASD group and a non-ASD group. The general data, surgical data, and preoperative spinal-pelvic sagittal imaging parameters [thoracic kyphosis (TK), lumbar lordosis (LL), segmental lordosis (SL), sacral slope (SS), pelvic tilt (PT), pelvic incident angle (PI), relative height of intervertebral space, etc] of the two groups were compared. Multivariate logistic regression analysis was used to analyze the correlations between preoperative spinal-pelvic sagittal imaging parameters and the occurrence of ASD after lumbar fusion surgery. A risk prediction model was constructed using R software, and the model's discrimination was evaluated through the receiver operating characteristic curve (ROC). The Bootstrap method was used for 1,000 repeated samplings to verify the predictive performance of the model. Statistical analysis was performed using independent sample t test and χ² test. Results　Of the 197 patients, 185 completed the follow-up (93.90%). Among them, 43 patients (23.24%) were in the ASD group, including 26 males and 17 females, with an age of (49.44±10.12) years old; 142 patients (76.75%) were in the non-ASD group, including 88 males and 54 females, with an age of (50.13±10.19) years old. Multivariate logistic regression analysis indicated that PT, SL, TK, PI-LL, and interference from adjacent segments during surgery were the influencing factors for ASD after lumbar fusion surgery (all P<0.05). The ROC analysis of the risk prediction model constructed based on these indicators showed that the area under the curve of the model to predict the postoperative ASD risk was 0.970, with a sensitivity of 97.7% and a specificity of 85.9%, indicating that the model had good discriminative ability. Bootstrap validation showed that the deviation calibration curve of the model was in good agreement with the ideal curve, with a mean absolute error (MAE) of 0.031. Conclusions　Spinal pelvic sagittal balance plays an important role in the growth and development of the spine, and PT, SL, TK, PI-LL, and interference from adjacent segments during surgery are influencing factors for ASD after lumbar fusion surgery. This study provides clinical evidences for predicting the risk of ASD after lumbar fusion surgery.

Key words:

Lumbar fusion surgery,  Adjacent segment disease,  Spinal pelvic parameters,  Intervertebral disc characteristics,  Risk model

摘要：

目的　通过分析脊柱-骨盆矢状面的测量，构建腰椎融合术后邻椎病（adjacent segment disease，ASD）的风险模型，并进行验证。方法　回顾性分析2020年10月至2023年10月在西安交通大学第一附属医院榆林医院接受经椎间孔入路椎间植骨融合术的197例患者资料。随访1年后，将最终纳入研究的患者分为ASD组和非ASD组。比较两组患者的一般资料、手术资料、术前脊柱-骨盆矢状面影像学参数［胸椎后凸角（TK）、腰椎前凸角（lumbar lordosis，LL）、节段性前凸（segmental lordosis，SL）、骶骨倾斜角（sacral slope，SS）、骨盆倾角（PT）、骨盆入射角（PI）、椎间隙相对高度等］，利用多因素logistic回归分析术前脊柱-骨盆矢状面影像学参数与腰椎融合术后ASD发生的相关性。运用R软件构建风险预测模型，并通过受试者操作特征曲线（ROC）评估模型的区分度。使用Bootstrap方法进行1 000次重复采样以验证模型的预测效能。采用独立样本t检验、χ²检验进行统计学分析。结果　纳入197例患者，最终185例完成随访（93.90%）。其中，ASD组43例（23.24%），包括男26例和女17例，年龄（49.44±10.12）岁；非ASD组142例（76.75%），包括男88例和女54例，年龄（50.13±10.19）岁。多因素logistic回归分析结果表明，PT、SL、TK、PI-LL和术中邻近节段的干扰是腰椎融合术后发生ASD的影响因素（均P<0.05）。基于这些指标构建的风险预测模型的ROC分析显示，模型预测术后ASD风险的曲线下面积为0.970，灵敏度为97.7%，特异度为85.9%，表明模型有较好的区分能力；Bootstrap验证表明，模型的偏差校准曲线与理想曲线吻合良好，均方误差（mean absolute error，MAE）为0.031。结论　脊柱-骨盆矢状面平衡在脊柱的生长发育中起着重要作用，而PT、SL、TK、PI-LL、术中邻近节段的干扰是腰椎融合术后发生ASD的影响因素，为预测腰椎融合术后ASD的风险提供了临床依据。

关键词:

腰椎融合术, 邻椎病, 脊柱-骨盆参数, 椎间盘特征, 风险模型