老年髋部骨折术后重症监护室患者医院获得性压力性损伤风险的网络计算器  

Network calculator for predicting the postoperative risk of hospital-acquired stress injuries of the elderly orthopaedic patients in the intensive care unit

作　　者：章晔 马菲菲 马少华 吴秋华 吴飞凤 王雪松 ZHANG Ye;MA Fei-fei;MA Shao-hua;WU Qiu-hua;WU Fei-feng;WANG Xue-song(Department of Intensive Care Unit,Rugao Hospital of Traditional Chinese Medicine,Rugao,Jiangsu,226500,China)

机构地区：[1]如皋市中医院重症监护室,江苏省226500 [2]如皋市中医院骨科,江苏省226500 [3]淮安市洪泽区人民医院重症监护室,江苏省223100

出　　处：《中国骨与关节杂志》2024年第12期1007-1013,共7页Chinese Journal of Bone and Joint

摘　　要：目的分析老年髋部骨折术后重症监护室(intensive care unit,ICU)患者医院获得性压力性损伤(hospital-acquired pressure injuries,HAPI)风险的危险因素,并构建列线图发布成网络计算器,为HAPI管理决策提供参考。方法选取我院2020年1月至2022年12月髋部骨折术后入ICU治疗的老年患者。采用多因素Logistic回归模型分析患者HAPI危险因素,将危险因素作为构建列线图预测指标。采用内部数据验证列线图性能,采用决策曲线分析评估列线图临床净收益,使用DynNom包将列线图发布至网络中并开发老年髋部骨折术后ICU患者HAPI的网络计算器。结果218例老年髋部骨折术后ICU患者HAPI发生率为19.3%(42/218)。多因素Logistic回归分析结果显示:年龄(OR:1.862,95%CI:1.267~4.563)、周围血管疾病(OR:1.750,95%CI:1.201~4.656)、Braden评分(OR:0.553,95%CI:0.238~0.863)、白蛋白(albumin,ALB)(OR:0.742,95%CI:0.538~0.964)和血乳酸(lactate,LAC)(OR:2.639,95%CI:1.265~5.684)是HAPI风险的独立危险因素(P<0.05)。列线图校正曲线显示C-index为0.837,当列线图预测HAPI风险阈值为0.097~0.802时,提供显著临床净收益。Simple模型预测1000例患者风险分层显示,风险阈值增高,列线图预测准确性逐渐接近实际观察结果。结论本研究基于年龄、周围血管疾病、Braden评分、ALB和LAC构建的老年髋部骨折术后ICU患者HAPI风险网络计算器,能早期、准确和便捷识别高风险患者,为实施更为积极干预措施提供依据。Objective To analyze the risk factors for hospital-acquired stress injury(HAPI)of the elderly orthopaedic patients in the intensive care unit(ICU)and construct a nomogram diagram to publish a web-based calculator for HAPI management decisions.Methods Elderly orthopedic patients who were admitted to the ICU for postoperative treatment between January 2020 and December 2022 in our hospital were selected.Patients’risk factors for HAPI were analyzed using multifactorial logistic regression models,and risk factors were used as predictors for constructing nomogram diagrams.Internal data were used to validate the performance of the nomogram,decision curve analysis was used to assess the net clinical benefit of the nomogram,and the DynNom package was used to publish the nomogram to the web and develop a web-based calculator for HAPI.Results The incidence of HAPI in 218 elderly ICU patients was 19.3%(42/218).Multifactorial logistic regression analysis showed that age(OR:1.862,95%CI:1.267-4.563),proportion of peripheral vascular disease(OR:1.750,95%CI:1.201-4.656),Braden score(OR:0.553,95%CI:0.238-0.863),ALB(OR:0.742,95%CI:0.538-0.964),and LAC(OR:2.639,95%CI:1.265-5.684)were HAPI independent risk factors(P<0.05).The results of receiver operating characteristic(ROC)curve analysis showed that the area under the ROC curve(AUC)for age,peripheral vascular disease,Braden score,ALB,and LAC were 0.666(0.599-0.728),0.606(0.537-0.671),0.707(0.642-0.767),0.755(0.692-0.811),and 0.815(0.757- 0.864 ), respectively;the nomogram correction curve showed a C-index of 0.837, which provided a significant clinicalbenefit when the nomogram predicted a risk threshold of 0.097 - 0.802 for HAPI. For Simple model prediction of 1000patients, risk stratification showed increasing risk thresholds and progressively closer nomogram prediction accuracyto actual observations. Conclusions In this study, a nomogram of HAPI risk in elderly postoperative hip fracture ICUpatients is constructed on the basis of age, peripheral vascular disease, Braden score,

关 键 词：重症监护病房 老年人 髋骨折 列线图 

分 类 号：R683.3[医药卫生—骨科学]