摘要
目的 探讨新生儿重症监护病房(NICU)新生儿医院感染临床特点和危险因素。方法 选取2019年6月-2020年6月医院新生儿病房收治的1 200例新生儿,其中发生医院感染70例为感染组,1 130例未发生医院感染为非感染组,多因素Logistic回归分析新生儿医院感染的危险因素,并建立风险预测模型,Hosmer-Lemeshow检验评估模型拟合度,受试者工作特征曲线(ROC)分析该模型的预测效能。结果 70例医院感染新生儿感染部位以肺部感染为主,占67.14%,其次为血液感染,占12.86%;共分离出85株病原菌,其中革兰阳性菌31株占36.47%,革兰阴性菌51株占60.00%,真菌3株占3.53%;多因素Logistic回归分析显示,基础疾病数量≥3种、有羊水污染和侵入性操作均为NICU新生儿医院感染的独立危险因素;建立的Logistic回归风险预测模型预测感染的曲线下面积(AUC)为0.754。结论 NICU新生儿医院感染病原菌以革兰阴性菌较为常见,Logistic回归风险预测模型可以较好地预测NICU新生儿医院感染的发生,临床需重点关注基础疾病数量≥3种、有羊水污染和侵入性操作的新生儿。
OBJECTIVE To explore the clinical characteristics and risk factors for nosocomial infection in neonates of neonatal intensive care unit(NICU). METHODS A total of 1200 neonates who were treated in neonatal wards from Jun 2019 to Jun 2020 were enrolled in the study, 70 of whom had nosocomial infection and were assigned as the infection group, 1130 did not have nosocomial infection and were assigned as the non-infection group. Multivariate logistic regression analysis was performed for the risk factors for nosocomial infection in the neonates, the risk prediction model was established, the degree of fitting of the model was evaluated by Hosmer-Lemeshow test, and the predictive efficiency of the model was analyzed by means of receiver operating characteristic(ROC) curves. RESULTS Among the 70 neonates with nosocomial infection, 67.14% had pulmonary infection, and 12.86% had blood infection. Totally 85 strains of pathogens were isolated, 31(36.47%) of which were gram-positive bacteria, 51(60.00%) were gram-negative bacteria, and 3(3.53%) were fungi. Multivariate logistic regression analysis showed that no less than 3 types of underlying diseases, amniotic fluid contamination and invasive operation were the independent risk factors for the nosocomial infection in the neonates of NICU;the area under curve(AUC) of the logistic regression risk prediction model was 0.754 in prediction of the infection. CONCLUSION The gram-negative bacteria are the common pathogens that are isolated from the NICU neonates with nosocomial infection. Logistic regression risk prediction model is effective in prediction of nosocomial infection in the neonates of NICU. It is necessary for the hospital to focus on the neonates with no less than 3 types of underlying diseases, amniotic fluid contamination and invasive operation.
作者
蒋琳
李欣妍
廖祯
陈琼英
曾颖
JIANG Lin;LI Xin-yan;LIAO Zhen;CHEN Qiong-ying;ZENG Ying(Longquanyi District Maternal and Child Health Hospital,Chengdu,Sichuan 61oioo,China)
出处
《中华医院感染学杂志》
CAS
CSCD
北大核心
2023年第1期125-128,共4页
Chinese Journal of Nosocomiology
基金
四川省科研基金项目(2020K02110)。
关键词
新生儿重症监护病房
医院感染
临床特点
相关因素
风险预测模型
Neonatal intensive care unit
Nosocomial infection
Clinical characteristic
Related factor
Risk prediction model
