通讯作者: 吴夏阳, 56425477@qq.com

DOI：10.12201/bmr.202505.00013

Establishment and verification of a risk prediction model for neonatal sepsis in premature infants

• 摘要：目的：旨在构建早产儿入院败血症风险预测模型,为临床早期识别及干预提供依据。方法：回顾性收集2020年1月至2023年12月厦门市儿童医院收治的早产患儿作为研究对象,根据入院后是否发生败血症分为败血症组和非败血症组,采用LASSO回归结合多因素Logistic回归筛选危险因素,构建列线图预测模型。选取2024年1月至2024年12月收治的174例患儿进行模型的外部验证。 结果：胎龄、10 min Apgar评分≤7、总胆红素、呼吸衰竭及呼吸频率是早产儿入院败血症的独立危险因素。模型训练集AUROC为0.853,外部验证AUROC为0.937,校准曲线中校正结果与理想曲线接近(Hosmer-Lemeshow检验χ2=6.599、P=0.580)。 结论：基于7项床旁指标构建的预测模型性能理想,无需微生物培养支持,可实现快速风险分层及抗菌药物决策。

  关键词： 早产儿; 败血症; 风险预测模型; 列线图; 危险因素

   

  Abstract: Objective: To construct a risk prediction model for admission sepsis in preterm infants, providing a basis for early clinical identification and intervention. Methods: A retrospective collection of preterm infants admitted to Xiamen Childrens Hospital from January 2020 to December 2023 was conducted as the study subjects. The infants were divided into a sepsis group and a non-sepsis group based on the occurrence of sepsis after admission. LASSO regression combined with multivariate Logistic regression was used to screen risk factors, and a nomogram prediction model was constructed. External validation of the model was performed with 174 infants admitted from January 2024 to December 2024. Results: Gestational age, Apgar score ≤7 at 10 minutes, total bilirubin, respiratory failure, and respiratory rate were identified as independent risk factors for admission sepsis in preterm infants. The AUROC of the training set was 0.853, and the external validation AUROC was 0.937. The calibration results in the calibration curve are close to the ideal curve (Hosmer-Lemeshow test χ2=6.599、P=0.580). Conclusion: The prediction model developed based on seven bedside indicators demonstrates excellent performance, enabling rapid risk stratification and antimicrobial decision-making without the need for microbiological culture support.

  Key words: Preterm infants; sepsis; risk prediction model; nomogram; risk factors

  提交时间：2025-05-17

  版权声明：作者本人独立拥有该论文的版权，预印本系统仅拥有论文的永久保存权利。任何人未经允许不得重复使用。
• html

• 图表

• 何永虹, 张伟, 覃大卫, 田文军, 陈玲, 严蜜, 谷秀, 符和建, 田昌军. 早产儿羊水胎粪污染的危险因素及临床结局. 2025. doi: 10.12201/bmr.202502.00007

  苏燕凤, 洪素茹, 陈钰霜, 吴夏阳. 结合机器学习模型的早产儿ICU死亡风险评估与可解释性分析. 2025. doi: 10.12201/bmr.202503.00066

  胡欣. 剖宫产术中寒战风险预测模型及列线图的构建. 2025. doi: 10.12201/bmr.202501.00053

  周妮, 谷强. 儿童大叶性肺炎病程的危险因素及风险预测. 2025. doi: 10.12201/bmr.202504.00059

  施明杰, 叶洪舟, 袁琛. 大环内酯类耐药肺炎支原体肺炎患儿发展为重症肺炎的危险因素分析及预测模型构建. 2024. doi: 10.12201/bmr.202409.00021

  陈冰融, 舒文秀, 罗柳飞, 童嘉琦, 乐静. 骨髓瘤患者发生3-4级骨髓瘤骨病的风险预测模型构建. 2024. doi: 10.12201/bmr.202410.00057

  张丽亚, 陈黎丽. ω－3PUFA对早产儿脑损伤免疫功能调节机制研究. 2024. doi: 10.12201/bmr.202408.00011

  谢芬、余娇通讯作者、宋思蓓、黄琳. 高甘油三酯血症性急性胰腺炎患者短期内复发再入院的风险预警模型构建与验证. 2024. doi: 10.12201/bmr.202408.00010

  王一凡, 石超君, 马安宁. Ⅱ型糖尿病并发动脉粥样硬化风险预测模型比较. 2024. doi: 10.12201/bmr.202404.00007

  何源, 杨树杰, 曾英, 易海英, 黄杰, 甘斌, 魏贤, 唐艳林. 微创给药在极早产儿呼吸窘迫综合征的疗效观察. 2024. doi: 10.12201/bmr.202410.00018

• 序号	提交日期	编号	操作
  1	2025-04-15	bmr.202505.00013V1	下载

• 公开评论  匿名评论  仅发给作者 提交评论