欢迎访问《中国循证儿科杂志》官方网站，今天是
English Version
中国循证儿科杂志

中国循证儿科杂志 ›› 2018, Vol. 13 ›› Issue (1): 40-44.

• 论著 • 上一篇    下一篇

重症监护病房非感染外科患儿体温、白细胞、C-反应蛋白和白介素-6对医院感染的预测价值

张晓磊,程晔,张铮铮,金爱丽,秦妍,沈伟杰,陆国平#br#   

  1. 复旦大学附属儿科医院重症医学科  上海，201102
  • 收稿日期:2017-11-15 修回日期:2018-02-25 出版日期:2018-02-25 发布日期:2018-02-25
  • 通讯作者: 陆国平

Value of body temperature, white blood cell count, C-reactive protein and interleukin-6 in the prediction of nosocomial infection in surgical patients without infection in PICU

ZHANG Xiao-lei, CHENG Ye, ZHANG Zheng-zheng, JIN Ai-li, QIN Yan, SHEN Wei-jie, LU Guo-ping   

  1. Pediatric Intensive Care Unit, Children's Hospital of Fudan University, Shanghai 201102, China
  • Received:2017-11-15 Revised:2018-02-25 Online:2018-02-25 Published:2018-02-25
  • Contact: LU Guo-ping

PDF

973

摘要： 目的：动态监测PICU非感染外科患儿体温（T）、WBC、 CRP和IL-6，探讨这些指标对发生医院感染（院感）的预测价值。 方法：选取2016年6月23日至2017年3月15日入住复旦大学附属儿科医院PICU且符合本文纳入、排除和剔除标准的外科急危重症患儿。根据是否发生院感分为院感组和非院感组；记录4个时点T、WBC、CRP和IL-6［P1为入住PICU后4 h内，P2~P4分别为P1后（48±1） h、（120±1）h和（192±1） h］，分析其对院感的预测价值。绘制受试者工作特征（ROC）曲线，计算曲线下面积（AUC）、灵敏度和特异度等。 结果：42例患儿进入本文分析，院感组11例，非院感组31例，两组在性别，入PICU时年龄、体重、T、WBC、CRP、IL-6，基础疾病分布和有创操作分布方面差异均无统计学意义（P＞0.05），两组中心静脉置管情况差异有统计学意义（P=0.03)。院感组与非院感组T、WBC、CRP和IL-6差异均有统计学意义（P<0.05），AUC分别为0.778、0.765、0.767和0.704，预测院感的最佳界值分别为T 37.3℃、WBC 10.3×109·L-1、CRP 27.0 mg·L-1和IL-6 55.0 pg·mL-1，敏感度63.6%~90.9%，特异度64.5%~83.9%。性别、年龄、入PICU时CRP、原发疾病、4种有创操作、气管插管、中心静脉置管和留置尿管均不是院感发生的独立危险因素。4项观察指标共有11种不同的组合，预测院感的AUC为0.754~0.842，敏感度72.7%~90.9%，特异度71.0％~96.8％。 结论：T+WBC+IL-6同时具备较高的敏感度和特异度，对院染的预测价值最好。

Abstract: Objective：Body temperature (T), white blood cell count, plasma C-reactive protein and interleukin-6 of non-infectious surgical patients in pediatric intensive care unit were monitored dynamically, to explore their value in predicting nosocomial infection. Methods：Patients were admitted to the study in Children's Hospital of Fudan University from June 23th, 2016 to March 15th, 2017, according to inclusion and exclusion criteria. They were categorized into two groups, nosocomial infection group and non-nosocomoal infection group, according to the diagnostic criteria of nosocomial infection. The values of T, WBC, CPR and IL-6 at 4 h (P1), (48±1)h (P2), (120±1)h (P3) and (192±1)h (P4) after admission were recorded and their value in predicting nosocomial infection was analyzed. Receiver-operating characteristic (ROC) curves were determined for them, and the areas under the curve (AUC), sensitivity and specificity were calculated. Results：A total of 42 patients were registered in the study, and 11 of them had nosocomial infection. There were no statistical difference for sex, age, weight, body temperature, WBC count, CRP, IL-6, primary disease distribution and invasive operation distribution between nosocomial infection group and non-nosocomoal infection group (P>0.05), but were statistical differences for central venous catheterization between the two groups (P=0.03), when patients were just admitted to PICU. The statistical differences of T, WBC, CRP and IL-6 were significant between nosocomial infection group and non-nosocomoal infection group (P<0.05), and the AUC of them was 0.778, 0.765, 0.767 and 0.704, respectively. The best cut-off values of them were 37.3 ℃, 10.3×109·L-1, 27.0 mg·L-1and 55.0 pg·mL-1, respectively.The sensitivity ranged from 63.6% to 90.9%, while the specificity ranged from 64.5% to 83.9%. Sex, age, CRP at admission, 4 kinds of invasive procedures, primary disease, endotracheal intubation, central venous catheterization and indwelling urethral catheter were not independent risk factors for nosocomial infection (P>0.05). The 4 indicators could be made up into 11 combinations, and the AUC ranged from 0.754 to 0.842, the sensitivity ranged from 72.7% to 90.9%, and the specificity ranged from 71.0% to 96.8%. Conclusion：The combination of T, WBC and IL-6 can be taken as an indicator for monitoring nosocomial infection with relatively high sensitivity and specificity.