术前颈椎椎体与终板骨质量评分预测颈椎前路融合术后融合器沉降风险效能的对比研究

A comparative study of the predictive efficacies of preoperative cervical vertebral bone quality score and endplate bone quality score on cage subsidence risk after anterior cervical discectomy and fusion surgery

中文关键词:  颈椎前路椎间盘切除融合术  颈椎终板骨质量评分  颈椎椎体骨质量评分  零切迹融合器  融合器沉降

英文关键词:Anterior cervical discectomy and fusion  Cervical endplate bone quality score  Cervical vertebral bone quality score  Zero-Profile  Cage subsidence

中文摘要:

  【摘要】 目的：评估术前颈椎椎体骨质量（cervical vertebral bone quality，CVBQ）和颈椎终板骨质量（cervical endplate bone quality，CEBQ）在预测单节段颈椎前路椎间盘切除融合术（anterior cervical discectomy and fusion，ACDF）后零切迹融合器（zero-profile，Zero-P）沉降中的应用价值。方法：回顾性分析在苏州大学附属第二医院采用零切迹融合器行ACDF的72例患者，收集患者一般资料（性别、年龄、并发症等）、实验室指标（血钙、血糖等）、手术相关信息（节段、融合器类型）和影像学资料（术前颈椎MRI和术后1周以内及3个月以上的颈椎正侧位X线片）。在术后1周以内及3个月以上的颈椎侧位X线片上测量椎体上下缘高度，两者之差即为融合器沉降距离，将沉降距离≥2mm的纳入沉降组，＜2mm的纳入未沉降组。在术前颈椎MRI矢状位T1加权像上，分别于手术相邻椎体的髓质区域、终板下骨区域及小脑延髓池脑脊液区划定感兴趣区（region of interest，ROI），计算手术相邻椎体髓质区域与终板下骨区域的平均信号强度与小脑延髓池脑脊液信号强度的比值，分别为CVBQ和CEBQ值。进一步按性别进行分层分析，明确CVBQ和CEBQ对沉降的预测价值是否受性别因素影响。采用Pearson相关系数分析CVBQ和CEBQ与沉降距离的关系，ROC曲线评估CVBQ和CEBQ对融合器下沉发生的诊断效能，先进行单因素Logistic回归分析筛选潜在危险因素，随后将其中具有统计学意义的项目，连同CVBQ及CEBQ一并纳入多因素分析，以评估其独立影响。结果：共14例（19.4%）患者发生融合器沉降。沉降组与非沉降组相比，年龄更大（男：61.4±16.5岁 vs 51.1±10.8岁，Ｐ=0.049；女：62.6±13.1岁 vs 51.2±7.8岁，Ｐ=0.005）、糖尿病比例更高（男：42.9% vs 3.4%，Ｐ=0.018；女：42.9% vs 6.9%，Ｐ=0.040）。沉降组CVBQ（男：2.75±0.73 vs 2.02±0.53，Ｐ=0.004；女：2.84±0.69 vs 2.00±0.40，Ｐ=0.005）和CEBQ（男：2.82±0.688 vs 2.05±0.56，Ｐ=0.004；女：2.94±0.68 vs 2.05±0.42，Ｐ=0.002）均显著性高于非沉降组。Pearson相关分析显示CVBQ（r=0.58，Ｐ<0.001）和CEBQ（r=0.59，Ｐ<0.001）均与沉降距离呈中等正相关。ROC曲线分析表明，CVBQ（AUC=0.83）和CEBQ（AUC=0.85）对沉降均有良好预测价值，且CEBQ的预测效能稍优。单因素Logistic回归分析显示，年龄、糖尿病、CEBQ及CVBQ均为沉降的危险因素（Ｐ<0.05）。多因素分析显示，在调整年龄和糖尿病后，CEBQ（OR=11.466，Ｐ=0.004）与CVBQ（OR=8.804，Ｐ=0.005）均为沉降的独立危险因素，年龄和糖尿病不是融合器沉降的独立危险因素，CEBQ与沉降的关联强度高于CVBQ。结论：CVBQ和CEBQ更高的患者融合器下沉的可能性更大，CEBQ和CVBQ都可以用于预测单节段ACDF术后患者融合器沉降，CEBQ比CVBQ预测能力更好。

英文摘要:

  【Abstract】 Objectives: To evaluate the clinical values of cervical vertebral bone quality(CVBQ) score and cervical endplate bone quality(CEBQ) score in predicting the settlement of Zero Profile(Zero-P) cage after anterior cervical discectomy and fusion(ACDF). Methods: A retrospective analysis was performed on 72 patients who underwent ACDF with Zero-P cage in the Second Affiliated Hospital of Soochow University. General patient information(including sex, age, comorbidity), laboratory parameters(such as blood calcium and blood glucose), surgical details(surgical level and cage type), and imaging data(preoperative cervical MRI and anteroposterior and lateral X-rays within one week and after more than three months) were collected. On the lateral X-rays taken within one week and >3 months postoperatively, the heights of the upper and lower margins of vertebrae were measured, and the difference between was defined as the cage subsidence distance. Patients with a subsidence distance≥2mm were assigned to the subsidence group, and those with <2mm subsidence distance were divided to the non-subsidence group. On preoperative cervical sagittal T1-weighted MRI, placing regions of interest(ROIs) in the medullary region of adjacent vertebral bodies, the subchondral bone region of the endplates, and the cerebrospinal fluid(CSF) in the cerebellomedullary cistern. CVBQ and CEBQ values were the ratios of the average signal intensity of the medullary region of the surgical adjacent vertebral bodies(for CVBQ) and the subchondral bone region of the endplates(for CEBQ) to the signal intensity of the cerebrospinal fluid in the cerebellomedullary cistern. A stratified analysis by gender was conducted to further clarify whether the predictive value of bone quality scores for subsidence is influenced by gender factors. The Pearson correlation coefficient was used to assess the relationship of CVBQ and CEBQ with cage subsidence distance. Receiver operating characteristic(ROC) curves were applied to evaluate the diagnostic efficacy of CVBQ and CEBQ for cage subsidence. Univariate logistic regression analysis was performed to screen for potential risk factors, after that age and diabetes, which showed statistical significance, along with CVBQ and CEBQ, were incorporated into the multivariate analysis to assess their independent effects. Results: Cage subsidence occurred in 14 patients(subsidence rate: 19.4%). Compared with the non-subsidence group, the subsidence group of patients was older(male: 61.4±16.5 years vs 51.1±10.8 years, P=0.049; female: 62.6±13.1 years vs 51.2±7.8 years, P=0.005) and had a higher proportion of patients with diabetes(male: 42.9% vs 3.4%, P=0.018; female: 42.9% vs 6.9%, P=0.040). Both CVBQ and CEBQ were significantly higher in the subsidence group(CVBQ-male: 2.75±0.73 vs 2.02±0.53, P=0.004; female: 2.84±0.69 vs 2.00±0.40, P=0.005; CEBQ-male: 2.82±0.688 vs 2.05±0.56, P=0.004; female: 2.94±0.68 vs 2.05±0.42, P=0.002). Pearson correlation analysis revealed a moderate positive correlation between both CVBQ(r=0.58, P<0.001) and CEBQ(r=0.59, P<0.001) with subsidence distance. ROC curve analysis indicated that both CVBQ(AUC=0.83) and CEBQ(AUC=0.85) had good predictive values for subsidence, with CEBQ showing slightly better predictive performance. Univariate logistic regression analysis identified age, diabetes, CEBQ, and CVBQ as risk factors for subsidence(P<0.05). Multivariate analysis showed that after adjusting for age and diabetes, both CEBQ(OR=11.466, P=0.004) and CVBQ(OR=8.804, P=0.005) remained independent risk factors for subsidence, while age and diabetes lost independent significance. The strength of association with subsidence was greater in CEBQ than in CVBQ. Conclusions: Patients with higher CVBQ and CEBQ scores are more likely to experience fusion cage subsidence, and both CEBQ and CVBQ can be used to predict cage subsidence in patients after single-level ACDF, CEBQ has better predictive ability than CVBQ.