老年全髋关节置换术患者隐性失血的风险模型建立和验证  

Establishment and validation of a risk model for hidden blood loss in elderly patients undergoing total hip arthroplasty

作　　者：杨泽龙[1] 蒋婷[1] 李伟[1] 李子泉[1] 杨飞 胡坤 Yang Zelong;Jiang Ting;Li Wei;Li Ziquan;Yang Fei;Hu Kun(Department of Orthopedics,The Second Clinical Medical College of North Sichuan Medical College·Nanchong Central Hospital,Nanchong Sichuan,637000,China)

机构地区：[1]川北医学院第二临床医学院·南充市中心医院骨科,四川南充637000

出　　处：《生物骨科材料与临床研究》2023年第3期54-58,63,共6页Orthopaedic Biomechanics Materials and Clinical Study

摘　　要：目的建立老年全髋关节置换术患者隐性失血的风险预测模型,并进行内部验证。方法选择2020年1月至2022年8月在南充市中心医院行全髋关节置换术的老年患者248例进行回顾性分析。观察可能影响老年全髋关节置换术患者隐性失血的相关因素,以术后隐性失血量480 mL为界值将患者分为两组。比较两组各可能影响老年全髋关节置换术患者隐性失血的相关因素,再以LASSO筛选出可能影响因素后行多因素Logistic回归,根据多因素分析结果建立列线图模型并进行验证。结果本研究纳入的行全髋关节置换术的老年患者248例中,高隐性失血量组39例,平均失血量为(582.17±76.18)mL;低隐性失血量组209例,平均失血量为(367.19±43.02)mL。经多因素Logistic回归分析结果显示:年龄、高血压、糖尿病、骨质疏松、手术时间、假体类型为老年全髋关节置换术患者隐性失血的独立影响因素(P<0.05)。根据多因素分析结果建立老年全髋关节置换术患者隐性失血预测列线图模型,ROC曲线分析结果显示,该列线图模型老年全髋关节置换术患者隐性失血的AUC为0.918[95%CI(0.880,0.955)],校准曲线结果显示,列线图模型预测老年全髋关节置换术患者高隐性失血的概率与实际概率基本吻合,H-L拟合优度检验结果显示,该列线图模型预测概率与实际概率比较,差异无统计学意义(P>0.05)。决策曲线分析结果显示:当该列线图模型预测老年全髋关节置换术患者隐性失血的概率阈值为0.15~0.95时,患者的净受益率大于0。结论老年全髋关节置换术患者隐性失血的发生与年龄、高血压、糖尿病等因素有关,根据上述因素建立的连线图模型用于预测老年全髋关节置换术患者高隐性失血有较高的准确度与区分度。Objective To establish and internally validate a risk prediction model for hidden blood loss in elderly patients undergoing total hip arthroplasty(THA).Methods A retrospective analysis was performed on 248 elderly patients who underwent total hip arthroplasty in Nanchong Central Hospital from January 2020 to August 2022.The relevant factors that may affect the occult blood loss of elderly patients with total hip arthroplasty were observed,and the patients were divided into two groups according to the threshold of 480 mL of occult blood loss after operation.The relevant factors that may affect the occult blood loss of the elderly patients with total hip arthroplasty were compared in the two groups,the LASSO was used to screen out the possible influencing factors and the multivariate Logistic regression was conducted.According to the results of the multivariate analysis,a nomogram model was constructed and verified.Results Among 248 elderly patients with total hip arthroplasty included in this study,39 patients were in the high occultic blood loss group,with an average blood loss of(582.17±76.18)mL;209 cases were in the low occultic blood loss group,with an average blood loss of(367.19±43.02)mL.The results of multivariate Logistic regression analysis showed that age,hypertension,diabetes,osteoporosis,operation time and type of prosthesis were risk factors of the recessive bleeding in elderly patients with total hip arthroplasty factor(P<0.05).According to the results of multivariate analysis,a nomogram model was established to predict occult bleeding in elderly patients undergoing total hip arthroplasty.ROC curve analysis showed that the AUC of this nomogram model for hidden blood loss in elderly patients undergoing total hip arthroplasty was 0.918[95%CI(0.880,0.955)].The results of the calibration curve showed that the probability of the nomogram model predicting high hidden blood loss in elderly patients undergoing total hip arthroplasty was basically consistent with the actual probability.The H-L goodness of f

关 键 词：老年 全髋关节置换 隐性失血 多因素分析 列线图模型 

分 类 号：R687.42[医药卫生—骨科学]