| 张莉,徐邦红,杨辉,等.脓毒症病儿肠内营养不耐受风险列线图模型的构建与验证[J].安徽医药,2025,29(10):2008-2012. |
| 脓毒症病儿肠内营养不耐受风险列线图模型的构建与验证 |
| Construction and validation of a column chart model for the risk of enteral nutrition intolerance in children with sepsis |
| |
| DOI：10.3969/j.issn.1009-6469.2025.10.019 |
| 中文关键词: 脓毒症 肠道营养 儿童 肠内营养不耐受 列线图 危险因素 预测 |
| 英文关键词: Sepsis Enteral nutrition Child Intestinal nutrition intolerance Column chart Risk factors Prediction |
| 基金项目:江苏省妇幼保健协会科研课题项目(FYX202033) |
|
| 摘要点击次数: 40 |
| 全文下载次数: 43 |
| 中文摘要: |
| 目的探讨构建与验证脓毒症病儿肠内营养不耐受的列线图模型。方法选取 2021年 4月至 2023年 4月南京医科大学附属儿童医院收治的 421例脓毒症病儿作为研究对象，按照 7∶3比例大致估算，采用随机数字表法分为建模组(295例)与验证组(126)，根据病儿是否发生肠内营养耐受情况将建模组病人分为耐受组(186例)和不耐受组(109例)。采用多因素 logistic回归分析脓毒症病儿的危险因素；采用 R软件和 rms程序包构建列线图模型，并评估模型预测效能。结果 295例脓毒症病儿有 109例发生肠内营养不耐受，发生率为 38.25%。耐受组和不耐受组在身体质量指数(BMI)、急性生理学和慢性健康状况评价Ⅱ(APACHEⅡ)评分、开始肠内营养时间、口服钾制剂、腹压、降钙素原(PCT)［(12.34±1.42)mg/L比(17.68±2.31)mg/L］、白蛋白［(14.67±3.02)g/L比(34.25±4.38)g/L］、C反应蛋白(CRP)［(38.14±3.89)mg/L比(58.34±5.62)mg/L］和前白蛋白［(64.35±10.32)mg/L比(129.67±13.57)mg/L］方面比较差异有统计学意义(P<0.05)。多因素 logistic回归分析 BMI≤16.8 kg/m2、APACHE Ⅱ评分>20分、开始肠内营养时间>48 h、口服钾制剂、腹压>15 mmHg、PCT、白蛋白、CRP和前白蛋白是脓毒症病儿发生肠内营养不耐受的危险因素(P<0.05)。建模组的验证结果显示，受试者操作特征曲线(ROC曲线)曲线下面积(AUC)为 0.90，区分度较好，H-L检验为 χ2=6.11，P=0.635，一致性良好。验证组 AUC为 0.96，区分度较好，H-L检验为 χ2=7.25，P=0.325，一致性良好。结论 BMI、APACHEⅡ评分、开始肠内营养时间、口服钾制剂、腹压、PCT、白蛋白、CRP和前白蛋白是脓毒症病儿肠内营养不耐受的危险因素，以此构建的列线图模型具有良好的区分度与一致性，能直观地预测脓毒症病儿发生肠内营养不耐受的风险。 |
| 英文摘要: |
| Objective To explore the construction and validation of a columnar graphical model of enteral nutrition intolerance in children with sepsis.Methods A total of 421 children with sepsis admitted to Children's Hospital of Nanjing Medical University fromApril 2021 to April 2023 were selected as the research subjects. According to the 7∶3 ratio, the children were randomly assigned into a modeling group (n=295) and a validation group (n=126) by random number table method. Patients in the modeling group were separat-ed into a tolerance group (n=186) and an intolerance group (n=109) based on whether the child developed tolerance to enteral nutrition.Risk factors for children with sepsis were analyzed using multivariate logistic regression; R software and the rms package were used toconstruct the column-line graphical model and to assess the model predictive efficacy.Results In this study, 109 out of 295 childrenwith sepsis developed enteral nutrition intolerance, with an incidence rate of 38.25%. There were statistical differences between the tol-erance and intolerance groups in terms of body mass index (BMI), acute physiology and chronic health evaluation Ⅱ (APACHE Ⅱ)score, start of enteral nutrition, oral potassium preparation, abdominal pressure, procalcitonin (PCT) [(12.34±1.42) mg/L vs. (17.68± 2.31) mg/L], albumin [(14.67±3.02) g/L vs. (34.25±4.38) g/L], C-reactive protein (CRP) [(38.14±3.89) mg/L vs. (58.34±5.62) mg/L], and prealbumin [(64.35±10.32) mg/L vs. (129.67±13.57) mg/L] (P<0.05). The results of multivariate Logistic regression analysis indicated that BMI ≤ 16.8 kg/m2, APACHEⅡ score>20 points, start of enteral nutrition>48 hours, oral potassium preparation, abdominal pressure>15 mmHg, PCT, albumin, CRP, and prealbumin were risk factors for enteral nutrition intolerance in children with sepsis (P<0.05). Thevalidation results of the modeling group showed that the area under the ROC curve (AUC) was 0.90, with good discrimination, and theH-L test results were χ2=6.11, P=0.635, indicating good consistency. The AUC of the validation group was 0.96, with good discrimina-tion, and the H-L test showed χ2=7.25, P=0.325, indicating good consistency.Conclusion BMI, APACHEⅡ score, start of enteral nu-trition, oral potassium preparation, abdominal pressure, PCT, albumin, CRP, and prealbumin are risk factors for enteral nutrition intol-erance in children with sepsis, and the column-line graphical model constructed in this way has good discriminatory power and consis-tency, and can intuitively predict the risk of developing enteral nutrition intolerance in children with sepsis. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
