经皮椎体成形术骨水泥分布评价及影响因素分析

Evaluation and influencing factor analysis of bone cement distribution after percutaneous vertebroplasty

中文关键词:  骨质疏松性椎体压缩性骨折  经皮椎体成形术  骨水泥分布  填充率

英文关键词:Osteoporotic vertebral compression fracture  Percutaneous vertebroplasty  Bone cement distribution  Filling rate

中文摘要:

  【摘要】 目的：探讨一种评价经皮椎体成形术（percutaneous vertebroplasty，PVP）骨水泥分布的新方法，并分析不同骨水泥分布的影响因素。方法：回顾性分析2015年5月~2018年5月在我科行PVP治疗的132例骨质疏松性椎体压缩性骨折（osteoporotic vertebral compression fracture，OVCF）患者，男23例，女109例，年龄77.33±8.27岁。根据术后正侧位X线片和不同层面CT图像上的骨水泥分布形态，计算骨水泥在椎体内的空间分布评分，将患者分为两组，A组评分为0~7分，B组评分为8~10分。两组患者术前、术后1d、术后3个月、术后6个月及术后12个月进行疼痛视觉模拟评分（visual analogue scale，VAS）；记录两组患者年龄、性别、术前骨密度值、骨折天数（骨折发生日期至手术日期），使用三维重建CT图像测量手术椎体体积、骨水泥体积并计算骨水泥填充率。比较两组患者上述临床资料及影像学参数差异以筛选骨水泥分布的潜在影响因素，多因素Logistic回归分析确定独立影响因素，构建受试者工作特征曲线（receiver operating characteristic curve，ROC），确定骨水泥填充率的临界值。结果：A组患者62例，B组患者70例。骨水泥分布情况：2分2例，3分8例，4分6例，5分8例，6分10例，7分28例，8分32例，9分22例，10分16例。两组患者术后1d、3个月、6个月及12个月的VAS评分与术前相比明显改善（P<0.05）。B组患者术后3个月VAS评分小于A组（P<0.05），术后1d、6个月及12个月的VAS评分两组间无显著性差异（P>0.05）。B组患者的骨折后手术时间明显小于A组，差异有统计学意义（P<0.05）；B组患者骨水泥体积、填充率均大于A组患者，差异有统计学意义（P<0.05）。多因素Logistic回归分析结果显示，骨水泥填充率是骨水泥分布的独立影响因素（OR=1.413；95%CI：1.202，1.660）。骨水泥填充率的ROC曲线下面积为0.778（95%CI：0.697~0.846，P<0.05），达到骨水泥广泛分布的临界值为28.1%，此时对应的敏感性和特异性分别为67.14%和75.81%。结论：PVP能够有效缓解OVCF患者疼痛，骨水泥分布范围随骨折时间延长而减小。骨水泥填充率是骨水泥分布的重要独立影响因素，当骨水泥填充率达到28.1%时即可实现广泛的骨水泥分布。

英文摘要:

  【Abstract】 Objectives: To explore a new method for evaluating the distribution of bone cement in percutaneous vertebroplasty(PVP), and then to analyze the influencing factors of bone cement distribution. Methods: From May 2015 to May 2018, a total of 132 patients(23 males and 109 females, mean age 77.33±8.27 years) with osteoporotic vertebral compression fracture(OVCF) who underwent PVP were recruited in this retrospective analysis study. The spatial distribution score of bone cement in the vertebral body was calculated according to postoperative X-ray and CT, and patients were allocated to two groups: group A: 0-7 points and group B: 8-10 points. The visual analogue scale(VAS) at preoperation, one-day, three-month, six-month and 12-month follow-up were compared between the two groups. The age, sex, preoperative bone mineral density, and number of fracture days (fracture date to surgery date) were recorded. Furthermore, the volume of the vertebral body, volume of bone cement and filling rate of bone cement were measured by using 3D reconstructed CT images. Potential influencing factors were screened after comparing the difference between the clinical and radiologic parameters in the two groups. Multivariate logistic regression analysis was performed to identify the independent influencing factors of bone cement distribution. Meanwhile, a receiver operating characteristic curve(ROC) was used to determine the cut-off value of the bone cement filling rate. Results: There were 62 patients in group A and 70 patients in group B. Distribution conditions of bone cement include: 2 points 2 cases, 3 points 8 cases, 4 points 6 cases, 5 points 8 cases, 6 points 10 cases, 7 points 28 cases, 8 points 32 cases, 9 points 22 cases, and 10 points 16 cases. VAS scores at one-day, three-month, six-month and 12-month follow-up improved significantly compared with preoperative ones in both groups(P<0.05). Group B had a better restoration of VAS scores at three-month follow-up than group A(P<0.05), but no significant difference was observed at one-day, six-month and 12-month follow-up. In group B, the number of days from fracture to surgery was significantly smaller than that of group A(P<0.05), while the bone cement volume and the filling rate were bigger than those in group A(P<0.05). Filling rate was identified as the independent risk factor influencing the distribution of bone cement by multivariate logistic regression analysis (OR=1.413, 95%CI: 1.202-1.660). The area under the receiver operating characteristic curve of filling rate was 0.778(95%CI: 0.697-0.846, P<0.05). The cut-off value of filling rate to reach a wide bone cement distribution was 28.1%, with a sensitivity of 67.14% and a specificity of 75.81% respectively. Conclutions: PVP can effectively relieve the pain of OVCF, and the bone cement distribution area gradually reduceswith the extension of fracture time. Filling rate was identified as the independent risk factor influencing the distribution of bone cement, and a wide bone cement distribution can be achieved when the filling rate reaches 28.1%.