摘要
目的分析胸椎黄韧带骨化症(TOLF)手术预后的影响因素,并构建预测模型。方法回顾性分析2016年2月至2019年2月重庆奉节县人民医院收治的138例行后路减压内固定手术的TOLF患者的临床资料,术后6个月对所有患者进行随访,评估预后效果,将疗效评判为"差""可"的患者纳入不良组(35例),将疗效评判为"良""优"的患者纳入优良组(103例),比较两组患者的临床资料,采用多因素Logistic回归分析影响行TOLF手术患者预后的危险因素,并构建预测模型,最后验证模型的效果。结果多因素Logistic回归分析结果显示,合并硬膜囊骨、T_(2)加权像髓内高信号、椎管面积残余率、术后脑脊液漏、病程是TOLF手术预后不良的危险因素(P<0.05)。预测模型:P=1/(e^^(-Y)),Y=2.419+2.726×合并硬膜囊骨+3.735×T_(2)加权像髓内高信号+3.616×椎管面积残余率+1.949×术后脑脊液漏+2.137×病程,曲线下面积为0.632,灵敏度为77.1%,特异度为75.7%,模型预测效果验证的灵敏度为84.21%、特异度为82.93%。结论合并硬膜囊骨、T_(2)加权像髓内高信号、椎管面积残余率、术后脑脊液漏、病程是TOLF手术预后不良的危险因素,采用由上述危险因素所构建的预测模型能够对TOLF手术预后不良的风险进行有效评估。
Objective To analyze the influencing factors of thoracic ossification of ligamentum flavum(TOLF)and to construct a predictive model.Methods The clinical data of 138 patients who underwent TOLF surgery during February 2016 to February 2019 in Fengjie County People′s Hospital were retrospectively analyzed.All patients were followed up 6 months after the operation to evaluate the prognosis.Patients whose efficacy evaluated as"fair"or"poor"were included in the poor prognosis group(35 patients),and those whose efficacy evaluated as"excellent""or"good"were included in the good prognosis group(103 patients).The clinical data of the two groups of patients were collected,and the risk factors affecting the prognosis of patients undergoing TOLF surgery were analyzed by multivariate Logistic regression method.A prediction model was constructed and it′s effect was verified finally.Results Multivariate Logistic regression analysis showed that complications of dural sac bone,T_(2)-weighted intramedullary high signal,spinal canal area residual rate,postoperative cerebrospinal fluid leakage and course of disease were risk factors for poor prognosis of TOLF surgery(P<0.05).Prediction model:P=1/(e^^(-Y)),Y=2.419+2.726×complications of dural sac bone+3.735×T_(2)-weighted intramedullary high signal+3.616×spinal canal area residual rate+1.949×postoperative cerebrospinal fluid leakage+2.137×course of disease.The area under the curve(AUC)was 0.632,the sensitivity and specificity were 77.1%and 75.7%,respectively.The sensitivity and specificity of model prediction validation were 84.21%and 82.93%respectively.Conclusions The complications of dural sac bone,T_(2)-weighted intramedullary high signal,spinal canal area residual rate,postoperative cerebrospinal fluid leakage and course of disease are risk factors for poor prognosis of TOLF surgery.Clinically,the prediction model constructed by the above risk factors is effective in assessing the risk of poor prognosis of TOLF surgery.
作者
马勇
陈伟男
Ma Yong;Chen Weinan(Department of Orthopedics,Fengjie County People′s Hospital,Chongqing Fengjie 404600,China)
出处
《中国医师进修杂志》
2021年第10期916-920,共5页
Chinese Journal of Postgraduates of Medicine
关键词
胸椎
黄韧带
骨化
预后
因素分析
统计学
Thoracic vertebrae
Ligamentum flavum
Ossification
Prognosis
Factor analysis,statistical
