From the Department of Pediatrics, Neonatal Division, McMaster University, Hamilton Health Sciences and McMaster Children Hospital, Hamilton, Canada.
Department of Pediatrics, Al-Baha University, Saudi Arabia.
Pediatr Infect Dis J. 2019 May;38(5):528-532. doi: 10.1097/INF.0000000000002182.
Neonatologists usually wait 48 hours for blood culture results before deciding to discontinue antibiotics. The objective of the study was to analyze time to positive blood culture in rule out sepsis and estimate the minimum duration of antibiotics.
Retrospective analysis of blood culture at the Neonatal Intensive Care Unit, McMaster Children's Hospital (January 2004 to December 2013) using BacT/Alert® 3D microbial system was conducted. We calculated average time taken for blood culture samples to emit a positive signal and compared it between Gram-positive and Gram-negative organisms. Kaplan-Meier curves for time to detect positive culture were generated. A Cox proportional hazard regression model with the outcome variable "time to detect positive blood culture" and predictor variables "early-onset sepsis (EOS) versus late-onset sepsis (LOS)", "Gram-positive versus Gram-negative" and "definite versus possible pathogen versus contaminant" was generated.
Of 7,480 blood cultures performed in 9,254 neonates, 885 samples grew microorganisms. 845 culture reports from 627 neonates were analyzed. Definite or opportunistic pathogens caused 815 (96%) infections (54 EOS and 791 LOS) and the rest were contaminants. Gram-negative organisms grew significantly faster than Gram-positive (P < 0.001). Cultures from EOS were positive significantly earlier than LOS (P = 0.032). Gram-negative status was an independent predictor of early detection of a positive culture (hazard ratio 3.5 [95% CI, 2.7-4.5] P < 0.001).
The probability of positive blood culture beyond 24 hours for a Gram-negative organism is small. Empiric antimicrobial treatment can be reduced after 24 hours to target only Gram-positive organisms in LOS and can be stopped in EOS unless clinical or laboratory parameters strongly suggest sepsis.
新生儿科医生通常会等待 48 小时，等待血培养结果，然后再决定是否停止使用抗生素。本研究旨在分析血培养排除败血症的阳性时间，并估计抗生素的最短持续时间。
对麦克马斯特儿童医院新生儿重症监护病房（2004 年 1 月至 2013 年 12 月）的血培养进行回顾性分析，使用 BacT/Alert® 3D 微生物系统。我们计算了血液培养样本发出阳性信号所需的平均时间，并比较了革兰氏阳性菌和革兰氏阴性菌之间的差异。绘制了时间至阳性培养的 Kaplan-Meier 曲线。使用以“阳性血培养时间”为因变量、“早发型败血症（EOS）与晚发型败血症（LOS）”、“革兰氏阳性菌与革兰氏阴性菌”、“明确病原体与可能病原体与污染菌”为预测变量的 Cox 比例风险回归模型。
在 9254 例新生儿中进行了 7480 次血培养，其中 885 次培养出微生物。对 627 例新生儿的 845 份培养报告进行了分析。明确或机会性病原体引起 815 例（96%）感染（54 例 EOS 和 791 例 LOS），其余为污染菌。革兰氏阴性菌的生长速度明显快于革兰氏阳性菌（P<0.001）。EOS 的培养阳性时间明显早于 LOS（P=0.032）。革兰氏阴性状态是早期检测阳性培养的独立预测因子（风险比 3.5[95%CI，2.7-4.5]，P<0.001）。
革兰氏阴性菌在 24 小时后血培养阳性的概率较小。对于 LOS 中的革兰氏阳性菌，经验性抗菌治疗可以在 24 小时后减少，而对于 EOS，除非临床或实验室参数强烈提示败血症，否则可以停止治疗。
