预测老年髋部骨折患者术后医院获得压力性损伤Nomogram模型  

Nomogram model for predicting hospital-acquired pressure injuries in elderly patients with hip fracture

作　　者：孟洁 张军[2] 葛绍勇 汪洋 李俊 MENG Jie;ZHANG Jun;GE Shao-yong;WANG Yang;LI Jun(Orthopaedic Nursing Department,Affiliated Hospital of Wanxi Health Vocational College,Lu'an,Anhui,237000,China)

机构地区：[1]皖西卫生职业学院附属医院骨科护理部,安徽237000 [2]皖西卫生职业学院附属医院骨科,安徽237000

出　　处：《中国骨与关节杂志》2022年第4期304-310,共7页Chinese Journal of Bone and Joint

基　　金：安徽高校自然科学研究重点项目(KJ2017A909)。

摘　　要：目的分析老年髋部骨折患者术后医院获得压力性损伤(hospital-acquired pressure injuries,HAPI)危险因素,并构建Nomogram模型,用于预测HAPI发生风险。方法选取2018年1月至2021年6月在我院手术治疗的315例老年髋部骨折患者。采用单因素和多因素Logistic回归模型分析HAPI危险因素。在Logistic回归模型基础上构建Nomogram模型。采用内部数据集进行模型的验证。结果(1)315例中58例(18.4%)观察到HAPI,术后至观察HAPI时间为2~5天,平均(2.84±1.15)天;(2)HAPI患者年龄、Charlson共病指数、美国麻醉医师协会(American Society of Anaesthesiologists,ASA)评分(3~4分)比例、营养风险筛查(nutritional risk screening-2002,NRS-2002)(营养不良)比例、Braden评分(≥15分)比例、C-反应蛋白(C-reactive protein,CRP)、血尿素氮(blood urea nitrogen,BUN)高于无HAPI患者(P<0.05),而BMI、血红蛋白(hemoglobin,Hb)、白蛋白(albumin,ALB)低于无HAPI患者(P<0.05);(3)年龄、Charlson共病指数、ASA评分(2~3分)、NRS-2002(营养不良)、Braden评分(≥15分)、CRP独立增加老年髋部骨折患者术后HAPI发生风险(P<0.05);(4)Nomogram模型预测风险截断值为0.44时,预测模型具有最佳特异度和灵敏度。内部数据集验证结果显示Nomogram模型一致性指数(consistency index,C-index)为0.724(95%CI:0.689~0.733)。决策曲线分析结果显示Nomogram模型预测HAPI发生风险的风险阈值在0.18~0.82时,提供显著附加标准化净收益。结论本研究构建的Nomogram模型可以有效预测老年髋部骨折患者术后HAPI发生风险。当预测风险>0.44时,该患者应当被视为HAPI高危患者。Objective To analyze the risk factors of hospital-acquired pressure injury(HAPI)in elderly patients with hip fracture,and construct Nomogram model to predict the risk of HAPI.Methods A series of315 elderly patients with hip fractures treated in our hospital from January 2018 to June 2021 were selected.Univariate and multivariate Logistic regression models were used to analyze the risk factors of HAPI.Nomogram model was constructed based on Logistic regression model.The internal data set was used to verify the Nomogram model.Results(1)HAPI was observed in 58(18.4%)of 315 patients.The time from operation to the observation of HAPI was2-5 days,with an average time of(2.84±1.15)days;(2)Age,Charlson comorbidity index,proportion of ASA score(3-4 points),proportion of NRS-2002(malnutrition),proportion of Braden score(≥15 points),CRP and BUN of HAPI patients were higher than those without HAPI(P<0.05),while BMI,Hb and ALB were lower than those without HAPI(P<0.05);(3)Age,Charlson comorbidity index,ASA score(2-3 points),NRS-2002(malnutrition),Braden score(≥15 points)and CRP independently increased the risk of postoperative HAPI in elderly patients with hip fracture(P<0.05);(4)When the cutoff value of Nomogram model was 0.44,the prediction model had the best specificity and sensitivity.The internal data set validation results showed that Nomogram model consistency index(C-index)was 0.724(95%CI:0.689-0.733).The results of decision curve analysis showed that Nomogram model provided significant additional standardized net income when the risk threshold of HAPI risk was 0.18-0.82.Conclusions Nomogram model constructed in this study can effectively predict the risk of postoperative HAPI in elderly patients with hip fracture.When the predicted risk is higher than 0.44,the patient should be considered as a high-risk patient of HAPI.

关 键 词：髋骨折 创伤和损伤 模型 统计学 

分 类 号：R683.4[医药卫生—骨科学]