椎体强化术治疗胸腰椎骨质疏松性椎体压缩性骨折早期疼痛缓解不佳的列线图预测模型构建  

Construction of nomogram model for poor early pain relief of thoracolumbar osteoporotic vertebral compression fractures treated with vertebral augmentation

作　　者：田龙 胡萌 卢俊松 龙俊任[1] Tian Long;Hu Meng;Lu Junsong;Long Junren(Department of Orthopaedics,Langzhong People’s Hospital,Nanchong 637400,Sichuan,China)

机构地区：[1]阆中市人民医院骨科,南充637400

出　　处：《脊柱外科杂志》2023年第5期331-337,共7页Journal of Spinal Surgery

摘　　要：目的建立椎体强化术治疗胸腰椎骨质疏松性椎体压缩性骨折(OVCF)早期疼痛缓解不佳的列线图预测模型。方法选择2020年1月—2022年4月采用椎体强化术治疗的OVCF患者228例,收集患者一般资料、基础疾病、疾病相关因素、手术相关因素等资料。根据术后早期疼痛缓解情况将患者分为疼痛缓解组(n=193)和疼痛缓解不佳组(n=35),比较2组患者上述资料的差异。采用LASSO回归方法筛选协变量,通过多因素logistic回归方法分析椎体强化术治疗OVCF早期疼痛缓解情况的影响因素,根据多因素分析结果建立列线图模型并进行内部验证。结果2组患者既往腰伤史、骨折严重程度、皮质断裂、椎体前缘高度恢复情况、椎体中间高度恢复情况、骨水泥渗漏、骨水泥分布类型、骨折椎体数、软组织封闭、CT值差异均有统计学意义(P<0.05)。LASSO回归和多因素logistic回归分析结果显示,既往腰伤史、骨折严重程度、皮质断裂、骨水泥-椎体百分比、椎体前缘高度恢复情况、骨水泥渗漏、骨水泥分布类型为椎体强化术治疗OVCF早期疼痛缓解不佳的独立影响因素(P<0.05)。采用上述因素建立列线图模型,受试者工作特征曲线分析结果显示,模型的曲线下面积为0.900(95%置信区间为0.824~0.976);Bootstrap验证显示,模型预测概率与实际概率基本吻合;H-L拟合优度检验显示,该列线模型预测概率与实际概率比较差异无统计学意义(P>0.05)。结论椎体强化术治疗OVCF早期疼痛缓解不佳主要受既往腰伤史、骨折严重程度、皮质断裂、骨水泥-椎体百分比、椎体前缘高度恢复情况、骨水泥渗漏、骨水泥分布类型等因素的影响,根据上述因素建立的列线图模型预测早期疼痛缓解不佳具有较好的区分度。Objective To establish a Nomogram model of poor early pain relief for thoracolumbar osteoporotic vertebral compression fractures(OVCF)treated with vertebral augmentation.Methods A total of 228 patients with OVCF who were treated with vertebral augmentation from January 2020 to April 2022 were selected for the study.The general information,underlying diseases,diseases-related factors and surgery-related factors of the patients were collected.According to the early postoperative pain relief,the patients were divided into pain relief group(n=193)and poor pain relief group(n=35),and the differences in the above data between the 2 groups were compared.LASSO regression was used to screen the covariates,and multivariate logistic regression was performed to select the factors that may affect the early pain relief of OVCF treated with vertebral augmentation.A nomogram model established based on the multivariate analysis results and internal verification was carried out.Results There were statistically significant differences in the history of lumbar injuries,fracture severity,cortical rupture,recovery of anterior vertebral height,recovery of middle vertebral height,bone cement leakage,distribution type of bone cement,number of fractured vertebral bodies,soft tissue sealing,and CT values between the pain relief group and the pain relief poor group(P<0.05).The results of multivariate analysis on the basis of LASSO regression showed that the history of lumbar injury,fracture severity,cortical rupture,percentage of vertebral body with bone cement,recovery of anterior vertebral height,bone cement leakage and distribution type of bone cement were independent influencing factors for poor early pain relief of OVCF treated with vertebral augmentation(P<0.05).The nomogram model was established based on the above factors.The analysis result of the receiver operating characteristic curve showed that the area under the curve of the model was 0.900(95%confidence interval was 0.824-0.976).Bootstrap verification showed that the predictio

关 键 词：胸椎 腰椎 骨折 压缩性 骨质疏松 椎体成形术 椎体后凸成形术 

分 类 号：R683.2[医药卫生—骨科学]