学龄期脓毒血症患儿高营养风险的影响因素分析及模型构建  

Analysis of influencing factors and model construction of high nutritional risk in school-age children with sepsis

作　　者：张冬平 孟利珠 俞伦倩 ZHANG Dongping;MENG Lizhu;YU Lunqian(Department of Paediatrics,Huai'an First People's Hospital,Huaian Jiangsu 223001,China)

机构地区：[1]淮安市第一人民医院儿科,江苏淮安223001 [2]淮安市第一人民医院胃肠外科,江苏淮安223001

出　　处：《中国急救复苏与灾害医学杂志》2024年第11期1446-1450,共5页China Journal of Emergency Resuscitation and Disaster Medicine

基　　金：江苏省卫生健康委员会医学科研立项项目(编号:H2019071)。

摘　　要：目的 探讨学龄期脓毒血症患儿高营养风险的影响因素,构建列线图预测模型,并对模型的预测价值进行验证。方法 采用简单随机抽样法选择2020年1月—2022年12月淮安市第一人民医院收治的400例脓毒血症患儿作为研究对象,另采用营养状况和生长风险筛查工具(STRONGkids)对患儿进行营养风险筛查,中低营养风险患儿纳入中低风险组,高营养风险患儿纳入高风险组,收集并比较两组患儿临床资料,进行单因素和多因素分析,采用二元Logistic回归分析评价影像患儿高营养风险的独立危险因素,采用Rstudio构建列线图预测模型,并采用五折交叉验证法和Hosmer-Lemeshow检验对该模型的预测价值进行验证。结果 单因素分析结果显示,疾病分型、机械通气、急性生理学和慢性健康评估系统Ⅱ(APACHEⅡ)评分、乳酸(LAC)、白蛋白(ALB)、降钙素原(PCT)均是学龄期脓毒血症患儿存在高营养风险的影响因素(P<0.05)。二元Logistic回归分析结果显示,疾病分型(严重脓毒症)、机械通气(是)、APACHEⅡ评分、LAC、PCT均为学龄期脓毒血症患儿高营养风险的独立危险因素,ALB为保护因素(P<0.05)。基于独立危险因素构建列线图预测模型,当总分>120分时,提示患儿存在高营养风险;当总分>170分时,提示患儿高营养风险发生概率为70%,当总分>188分时,提示患儿高营养风险发生概率超过90%。五折交叉验证结果显示,训练集与测试集模型预测患儿高营养风险的曲线下面积(AUC)分别为0.903(95%CI=0.866~0.939)和0.868(95%CI=0.791~0.946);Hosmer-Lemeshow检验χ2=9.869,P=0.274。结论 多种因素均可导致学龄期脓毒血症患儿出现高营养风险,基于各项独立危险因素和保护因素构建列线图预测模型,有助于临床对脓毒血症患儿是否存在高营养风险作出准确判断,以便于制定个性化早期营养方案,改善患儿预后。Objective To explore the influencing factors of high nutritional risk in school-age children with sepsis,and to construct a nomogram prediction model and verify the predictive value of the model.Methods A simple random sampling method was used to select 400 children with sepsis admitted to Huai 'an First People 's Hospital from January 2020 to December 2022 as the study subjects.STRONGkids was used to screen the nutritional risk of the children.Children with low nutritional risk were included in the low-risk group,and children with high nutritional risk were included in the high-risk group.The clinical data of the two groups were collected and compared,and univariate and multivariate analysis were performed.Binary Logistic regression analysis was used to evaluate the independent risk factors of high nutritional risk in children with imaging.Rstudio was used to construct a nomogram prediction model.Five-fold cross-validation and Hosmer-Lemeshow test were used to verify the predictive value of the model.Results The results of univariate analysis showed that disease classification,mechanical ventilation,acute physiology and chronic health evaluation Ⅱ(APACHE Ⅱ) score,lactic acid(LAC),albumin(ALB),procalcitonin(PCT) were the influencing factors of high nutritional risk in school-age children with sepsis(P<0.05).A nomogram prediction model was constructed based on independent risk factors.When the total score was > 120,it indicated that the children had high nutritional risk.When the total score was > 170,it indicated that the probability of high nutritional risk was 70%.When the total score was > 188,it indicated that the probability of high nutritional risk was more than 90%.The results of the five-fold cross-validation showed that the AUCs of the training set and test set models for predicting the risk of high nutrition in children were 0.903(95% CI:0.866-0.939) and 0.868(95% CI:0.791-0.946),respectively;HosmerLemeshow test χ~2=9.869,P=0.274.Conclusion A variety of factors can lead to high nutritional risk in

关 键 词：脓毒血症 学龄期患儿 高营养风险 列线图 

分 类 号：R473.72[医药卫生—护理学]