摘要
目的调查老年股骨颈骨折手术病人1年死亡率,并构建其预测模型。方法选择2019年2月至2020年5月在我院接受手术治疗的老年股骨颈骨折病人208例进行回顾性分析。收集可能影响病人1年内死亡的相关临床资料。根据病人1年内生存情况将其分为生存组与死亡组,比较2组病人临床资料,并采用Logistic回归分析老年股骨颈骨折病人术后1年死亡的影响因素,以多因素分析结果构建列线图预测模型。结果生存组和死亡组年龄、入院前合并症、骨折前Harris髋关节评分、美国卒中协会(ASA)分级、骨折类型、WBC、Hb、白蛋白、活化部分凝血活酶时间差异均有统计学意义(P<0.05)。多因素Logistic回归分析结果显示:入院前合并症≥2种、ASA分级Ⅲ~Ⅳ级、囊内骨折为老年股骨颈骨折病人术后1年死亡的危险因素,Hb和骨折前Harris髋关节评分为保护因素(P<0.05)。以各因素β值构建老年股骨颈骨折病人术后1年内死亡预测模型:Prob=1/(1+e-Y),Y=0.389×入院前合并症-0.385×骨折前Harris髋关节评分+0.371×ASA分级+0.415×骨折类型-0.386×Hb,以此构建列线图模型。绘制ROC曲线显示,预测模型的AUC为0.964(95%CI:0.936~0.992),Bootstrap法内部验证结果显示:平均绝对误差为0.016,模型表现与理想模型基本拟合。结论老年股骨颈骨折病人术后1年死亡率主要受合并症、ASA分级、骨折类型、Hb和髋关节功能等因素的影响,以上述因素构建的列线图模型可有效预测老年股骨颈骨折病人术后1年死亡风险,模型具有较高的区分度与准确度。
Objective To investigate the 1-yeaRmortality of the elderly patients with femoral neck fracture receiving surgery,and to construct a predictive model.Methods A total of 208 elderly patients with femoral neck fractures who were treated with surgery in ouRhospital from February 2019 to May 2020 were enrolled in this study.The clinical data were collected foRretrospective analysis.According to the survival of the patients,they were divided into the survival group and the death group.The clinical data of the two groups were compared,and Logistic regression analysis was used to evaluate the influencing factors of 1-yeaRsurvival.A nomogram prediction model was constructed based on the results of the multivariate analysis.Results There were significant difference in age,comorbidities before admission,pre-fracture Harris hip score,American stroke association(ASA)grade,fracture type,the levels of white blood cell(WBC),hemoglobin(Hb),albumin and activated partial thromboplastin time between the survival group and the death group(P<0.05).Multivariate Logistic regression analysis showed that≥2 kinds of comorbidities before admission,ASA gradeⅢ-Ⅳ,and intracapsulaRfracture were the risk factors foR1-yeaRdeath in the elderly patients with femoral neck fracture.Increased hemoglobin and pre-fracture Harris hip score were the protective factors(P<0.05).Theβvalue of each factoRwas used to construct a one-yeaRmortality prediction model foRthe elderly patients with femoral neck fracture:Prob=1/(1+e-Y),Y=0.389×pre-hospital comorbidities-0.385×pre-fracture Harris hip score+0.371×ASA grade+0.415×fracture type-0.386×Hb.Nomogram model was constructed based on this formula.The receiveRoperating characteristic(ROC)curve of the model was drawn,and the area undeRthe ROC curve was 0.964(95%CI:0.936-0.992).The internal verification results of the Bootstrap method showed that the mean absolute erroRwas 0.016,and the model performance fitted the ideal model.Conclusions The 1-yeaRmortality of the elderly patients with femoral neck frac
作者
卢俊松
吕国荣
田龙
何涛
严纯
陈路
杨志强
LU Jun-song;LYU Guo-rong;TIAN Long;HE Tao;YAN Chun;CHEN Lu;YANG Zhi-Qiang(Department of Orthopedics,Langzhong People’s Hospital,Langzhong 637400,China;Department of Orthopedics,Affiliated Hospital of North Sichuan Medical College,Nanchong 637000,China)
出处
《实用老年医学》
CAS
2023年第2期151-155,共5页
Practical Geriatrics
基金
四川省医学会静脉栓塞症防治(恒瑞专项)科研课题(2019HR19)。
关键词
老年人
股骨颈骨折
预后
列线图
预测模型
aged
femoral neck fracture
prognosis
nomogram
predictive model
