糖尿病足溃疡复发风险预测模型的系统评价  

Risk prediction models for recurrence of diabetic foot ulcers:a systematic review

作　　者：谢玉生 黄蓉蓉 罗玉红 王乾沙 明玥 徐毅 Xie Yusheng;Huang Rongrong;Luo Yuhong;Wang Qiansha;Ming Yue;Xu Yi(School of Nursing,Guizhou Medical University,Guiyang 550025,China;Department of Burn and Plastic Surgery,the Affiliated Hospital of Guizhou Medical University,Guiyang 550001,China;School of Nursing,Capital Medical University,Beijing 100069,China;Department of Orthopedics,the Affiliated Hospital of Guizhou Medical University,Guiyang 550001,China;Department of Endocrine and Metabolism,the Affiliated Hospital of Guizhou Medical University,Guiyang 550001,China)

机构地区：[1]贵州医科大学护理学院,贵阳550025 [2]贵州医科大学附属医院烧伤整形外科,贵阳550001 [3]首都医科大学护理学院,北京100069 [4]贵州医科大学附属医院骨科,贵阳550001 [5]贵州医科大学附属医院内分泌代谢科,贵阳550001

出　　处：《中华现代护理杂志》2024年第11期1414-1421,共8页Chinese Journal of Modern Nursing

基　　金：贵州省卫生健康委科学技术基金项目(gzwkj2024-541);贵州医科大学护理学科专项科研项目(YJ22029)。

摘　　要：目的系统评价糖尿病足溃疡(DFU)复发风险预测模型。方法计算机检索PubMed、Embase、Web of Science、Cochrane Library、中国知网、万方和中国生物医学文献服务系统中有关DFU复发风险预测模型的研究,检索时限均为建库至2023年7月20日。2名研究员独立筛选文献,并依据预测模型研究数据提取表和偏倚风险评估工具(PROBAST)进行资料提取和质量评价。结果共纳入8篇文献,包含14个模型。纳入模型的ROC曲线下面积为0.660~0.940。模型中最常见的前5个预测因子依次为溃疡位置、糖化血红蛋白、吸烟、合并周围神经病变和糖尿病病程。PROBAST结果显示,8篇文献均存在较高的偏倚风险,主要原因是样本量不足、缺失数据的处理和报告不当、缺乏内部验证可能导致的模型过度拟合,且仅有1篇文献进行外部验证。结论DFU复发风险预测模型的研究尚处于发展阶段,各研究预测性能尚可,但存在较高的偏倚风险,未来研究仍需采用严谨的统计分析方法构建新的风险预测模型,并完善内部及外部验证。Objective To systematically evaluate the recurrence risk prediction model of diabetic foot ulcers(DFU).Methods Research on DFU recurrence risk prediction models was electronically searched in PubMed,Embase,Web of Science,Cochrane Library,China National Knowledge Infrastructure,Wanfang Data,and China Biomedical Service System.The search period was from database establishment to July 20,2023.Two researchers independently screened literature and conducted data extraction and quality evaluation using the prediction model research data extraction table and the Prediction Model Risk of Bias Assessment Tool(PROBAST).Results A total of 8 articles were included,including 14 models.The area under the receiver operating characteristic(ROC)curve included in the model ranged from 0.660 to 0.940.The most common five predictors in the model were ulcers location,glycosylated hemoglobin,smoking,combined peripheral neuropathy and diabetes course.All 8 articles had a high risk of bias,mainly due to insufficient sample size,improper handling and reporting of missing data,and a lack of internal validation,which might lead to overfitting of the model.Only one article was subjected to external validation.Conclusions The research on DFU recurrence risk prediction models is still in the development stage,and the predictive performance of various studies is still acceptable,but there is a high risk of bias.Future research still needs to use rigorous statistical analysis methods to construct new risk prediction models and improve internal and external validation.

关 键 词：糖尿病足 复发 预测模型 系统评价 

分 类 号：R473.5[医药卫生—护理学]