经皮椎体成形术后新发椎体压缩骨折临床预测模型的建立与验证

目的 探讨经皮椎体成形术(percutaneous vertebroplasty, PVP)术后新发椎体压缩性骨折（vertebral compression fractures ,VCFs)的危险因素并建立与验证预测模型。方法 选取2016年6月至2018年6月，在柳州市人民医院脊柱外科接受PVP治疗并符合本研究纳入标准的骨质疏松性VCFs (osteoporotic VCFs，OVCFs)患者进行回顾性研究。观察指标为年龄、性别、骨密度 (bone mineral density, BMD)、身高、体重、体质量指数 (body mass index, BMI)、住院时间、骨水泥量、骨水泥是否渗漏、手术时间、住院到手术时间、受伤到手术时间、术后是否进行抗骨质疏松症治疗、是否多椎体骨折以及类固醇药物使用对PVP术后新的压缩性骨折的发生情况进行单因素和多因素分析，以确定相关的危险因素，并建立预测模型并进行验证。结果 共有385名患者符合纳入标准，其中女性308例，男性77例。随访时间24~36个月，平均26.4个月。在385例患者中观察到58例新发的OVCFs。统计学分析显示，较高的BMI(P<0.01)、较低BMD（P<0.01）、多椎体骨折(P<0.05)、未进行抗骨质疏松症治疗（P<0.05）与使用类固醇药物（P<0.001）是导致PVP术后新发OVCFs的独立危险因素。最明显的是，使用类固醇药物令新骨折的风险增加了4.07倍(95 % CI为2.005~8.264)。并进一步建立了临床预测模型(Nomogram)及其验证，其内部ROC=0.796；验证组进行外部验证ROC=0.648，提示该模型具有较好预测能力。结论 BMI、BMD、多椎体骨折、未进行抗骨质疏松症治疗与使用类固醇药物是PVP术后新发OVCFs高危因素。

Risk factors analysis of new vertebral compression fractures after percutaneous vertebroplasty and establishment and validation of clinical prediction model

Objective To investigate the risk factors of new vertebral compression fracture (VCFs) after percutaneous vertebroplasty (PVP) and to establish and verify the prediction model. Methods From June 2016 to June 2018, background patients with osteoporotic VCFs (OVCFs) who received PVP treatment in the Spinal Surgery department of Liuzhou People''s Hospital and follow the inclusion criteria in this study were reviewed. Observation indexes for age, sex, bone mineral density, height, weight, body mass index, length of hospital stay, whether bone cement content, bone cement leakage, operation time, hospitalization to operation time, injury to the operation time, whether for anti-osteoporosis treatment after operation, whether more vertebral fracture and steroid use for PVP new compression fractures after the happening of single factor and multiple factors analysis, to determine the relevant risk factors, and established the forecast model and verified. Results A total of 385 patients met the inclusion criteria. There were 308 females and 77 males. The follow-up time was 24-36 months, 26.4 months on average. Fifty-eight new cases of OVCFs were observed in 385 patients. Statistical analysis showed that higher BMI, lower BMD, multiple vertebral fractures, lack of anti-osteoporosis therapy and use of steroid drugs were independent risk factors for new OVCF after PVP. Most notably, steroid use was associated with a 4.07-fold increased risk of new fractures (95% CI: 2.005-8.264).The clinical prediction model (Nomogram) was further established and validated, with internal ROC=0.796.The validation group carried out external validation ROC=0.648, suggesting that the model has good predictive ability. Conclusion BMI, BMD, multiple vertebral fractures, no anti-osteoporosis therapy and steroid use are the risk factors for new OVCFs after PVP.