老年髋部骨折患者术后1a内全因死亡预测模型的构建  

Construction of a prediction model for all-cause mortality within 1 year after hip fracture surgery in elderly patients

作　　者：靳永俊 肖鹏[1] 朱旭[1] 赵冰涛 吴学建[1] JIN Yongjun;XIAO Peng;ZHU Xu;ZHAO Bingtao;WU Xuejian(Department of Orthopedics,the First Affiliated Hospital,Zhengzhou University,Zhengzhou 450052)

机构地区：[1]郑州大学第一附属医院骨外科,郑州450052

出　　处：《郑州大学学报（医学版）》2024年第6期863-867,共5页Journal of Zhengzhou University(Medical Sciences)

摘　　要：目的:探讨老年髋部骨折患者术后1 a内全因死亡的影响因素并构建预测模型。方法:选择郑州大学第一附属医院收治的505例老年髋部骨折患者,按照4∶1的比例分为训练集(n=404)和验证集(n=101)。在训练集中采用Logistic回归分析筛选患者术后1 a内全因死亡的危险因素,并构建列线图预测模型。分别在训练集和验证集中对模型的区分度、校准度和临床效益进行评价。结果:高龄、男性、低血红蛋白、高血尿素氮、低血白蛋白是老年髋部骨折患者术后1 a内全因死亡的危险因素,OR(95%CI)分别为1.077(1.037~1.118)、2.360(1.269~4.389)、0.983(0.967~0.999)、1.108(1.039~1.181)、0.926(0.858~1.000)。训练集及验证集列线图预测模型的AUC(95%CI)为0.795(0.741~0.850)、0.796(0.658~0.934),提示模型的区分度较好。训练集及验证集的Hosmer-Lemeshow检验示模型拟合效果良好(P>0.05),校准曲线与理想曲线接近,提示模型的校准度较好。决策曲线显示模型具有较好的临床收益。结论:构建的列线图预测模型对老年髋部骨折术后1 a全因死亡的预测效能较好。Aim:To explore the factors influencing all-cause mortality within 1 year after hip fracture surgery in elderly patients and construct a prediction model.Methods:A total of 505 elderly hip fracture patients admitted to the First Affiliated Hospital of Zhengzhou University were selected.These patients were allocated into training set(n=404)and validation set(n=101)according to the ratio of 4∶1.Logistic regression analysis was conducted using the data of training set to identify the risk factors for all-cause mortality within 1 year after hip fracture surgery and a nomogram predictive model was constructed.The discrimination,calibration,and clinical benefit of the model were evaluated separately in the training and validation sets.Results:Advanced age,male,reduced hemoglobin,elevated blood urea nitrogen,and reduced blood albumin were risk factors for all-cause mortality within 1 year after hip fracture surgery in elderly patients,with OR(95%CI)of 1.077(1.037-1.118),2.360(1.269-4.389),0.983(0.967-0.999),1.108(1.039-1.181),0.926(0.858-1.000),respectively.The AUC(95%CI)for the training and validation sets was 0.795(0.741-0.850)and 0.796(0.658-0.934),respectively,indicating good discrimination of the model.Hosmer-Lemeshow test for the training and validation sets showed that the model fit well(P>0.05),and the calibration curve was close to the ideal curve,indicating that the model had good calibration.The decision curve analysis showed that the model exhibited a promising clinical benefit.Conclusion:The constructed nomogram prediction model shows good predictive performance for all-cause mortality within 1 year after hip fracture surgery in the elderly.

关 键 词：髋部骨折 全因死亡 预测模型 老年人 

分 类 号：R683.42[医药卫生—骨科学]