心脏外科术后患者皮肤撕裂伤风险预测模型的构建与验证  

Construction and validation of a risk prediction model for skin tear in postoperative cardiac surgery patients

作　　者：管艳 王哲芸[1] 金花[1] 陈晨 张驰 苏云艳[1] 余云 Guan Yan;Wang Zheyun;Jin Hua;Chen Chen;Zhang Chi;Su Yunyan;Yu Yun(Department of Cardiac Surgery,Nanjing Drum Tower Hospital,Nanjing University Medical School,Nanjing 210008,China)

机构地区：[1]南京大学医学院附属鼓楼医院心脏外科,江苏南京210008

出　　处：《护理学杂志》2024年第24期25-30,共6页Journal of Nursing Science

基　　金：江苏省“十四五”医学重点学科项目(ZDXK202229);南京大学中国医院改革发展研究院课题(NDYGN2023020,NDYGN2023050);南京鼓楼医院护理部课题(2023-B1006)。

摘　　要：目的构建并验证心脏外科患者皮肤撕裂伤风险预测模型,为制订预防性干预方案提供参考。方法采用前瞻性研究设计,选取2022年6月至2023年12月750例心脏外科术后患者作为研究对象。纳入70%患者作为建模组(525例),30%患者作为验证组(225例)。采用单因素和多因素logistic回归分析心脏术后患者皮肤撕裂伤的危险因素,建立风险预测模型并构建列线图,并对该模型的预测效果进行验证。结果建模组有122例发生皮肤撕裂伤,发生率23.2%。回归分析结果显示,年龄、糖尿病史、皮肤撕裂伤史、淤斑、低氧血症、失禁相关性皮炎、谵妄、类固醇药物使用数量是心脏术后患者发生皮肤撕裂伤的独立影响因素(均P<0.05)。以上述8个因素作为自变量绘制列线图,模型的ROC曲线下面积为0.807,灵敏度为0.691,特异度为0.802。模型在验证组测试的ROC曲线下面积为0.801,灵敏度为0.776,特异度为0.705。结论构建的模型预测效果良好,可为临床医护人员早期快速识别心脏术后患者发生皮肤撕裂伤风险、及时提供预防性干预方案提供参考。Objective To construct and validate a risk prediction model for skin tears in cardiac surgery patients,and provide a reference for developing preventive intervention programs.Methods A prospective research design was adopted,with 750 postope-rative cardiac surgery patients selected from June 2022 to December 2023 as study subjects.Seventy percent of the patients were included in the modeling group(525 cases),and thirty percent were included in the validation group.Univariate and multivariate logistic regression analyses were conducted to identify the risk factors for skin tears in postoperative cardiac patients,and a risk prediction model along with a nomogram was established.The predictive performance of the model was subsequently validated.Results In the modeling group,122 cases experienced skin tears,resulting in an incidence rate of 23.2%.Regression analysis indicated that age,history of diabetes,history of skin tears,ecchymosis,hypoxemia,incontinence-associated dermatitis,delirium,and the number of corticosteroid uses were independent risk factors for skin tears in postoperative cardiac patients(all P<0.05).A nomogram was created using these eight factors as independent variables,with the area under the ROC curve for the model being 0.807,sensitivity 0.691,and specificity 0.802.In the validation group,the ROC curve area was 0.801,with sensitivity at 0.776 and specificity at 0.705.Conclusion The constructed model demonstrates good predictive performance,providing clinical healthcare professionals with a reference for early and rapid identification of the risk of skin tears in postoperative cardiac patients and timely implementation of preventive intervention programs.

关 键 词：心脏手术 重症监护病房 皮肤撕裂伤 预测模型 危险因素 风险评估 心外科护理 

分 类 号：R473.6[医药卫生—护理学]