Late-onset sepsis (LOS) remains a major cause of mortality and morbidity among preterm infants. Blood culture is recognized as the gold standard of sepsis diagnosis. However, the sensitivity of blood culture is suboptimal in neonates because of low colony bacteremia, limited blood volume and inappropriate sampling;1,2 and a negative culture result does not totally exclude sepsis.3,4 It happens frequently that infants exhibit clinical signs and symptoms of sepsis but negative blood cultures.4,5 Many infants in such scenarios will be diagnosed with culture-negative sepsis and antibiotics will be commenced.5 Most reports describing the burden or outcomes of LOS focused solely on culture-positive LOS and excluded culture-negative LOS.6–9 However, culture-negative LOS is commonly diagnosed in clinical practice and may actually be more common than culture-positive LOS. Currently, there is a paucity of published data on culture-negative LOS and there are several important questions to be addressed. First, no large-scale observational study has reported the incidence of culture-negative LOS in neonatal intensive care units (NICUs). A trial of enteral lactoferrin supplement for prevention of neonatal LOS reported that 12.8% of infants <32 weeks of gestation developed culture-negative LOS;10 however, the relatively strict inclusion and exclusion criteria of clinical trials may affect the generalizability of the study population to the general population. Second, the influence of culture-negative sepsis on neonatal outcomes remains unclear. One large follow-up study showed culture-negative sepsis was associated with poor neurodevelopmental and growth outcomes in early childhood in a large cohort of infants with birth weight <1000 g,11 whereas this association was not observed in another cohort.12 Third, there remains controversy as to whether culture-negative sepsis is a clinically significant entity and whether antibiotics should be continued when the culture results were negative.13 Neonates with culture-negative sepsis contribute significantly to antibiotic consumption in NICUs. One study showed antibiotic courses targeting culture-negative sepsis was 3.7 times more than those targeting proven infections.14 There have been evolving concerns about the unintended harm from prolonged antibiotic exposure in culture-negative sepsis, which may include increased risk for necrotizing enterocolitis, bronchopulmonary dysplasia (BPD), severe neurologic injury, retinopathy of prematurity (ROP) and death.15,16 Therefore, some authors suggested antibiotics should be stopped once blood culture results are available,13 while others considered that continuing antibiotics might be reasonable in infants with other evidence highly suggestive of infection, but with limited duration.17
Therefore, the objectives of this study were to investigate the incidence of culture-negative LOS in a multicenter cohort of preterm infants in China and to examine neonatal outcomes among infants with culture-negative LOS compared with those without LOS or with culture-positive LOS.
MATERIALS AND METHODS
Twenty-five tertiary hospitals from 19 provinces in China participated in a cluster-randomized controlled study entitled “Reduction of Infection in Neonatal Intensive Care Units using the Evidence-based Practice for Improving Quality” (REIN-EPIQ study, clinicaltrials.gov #NCT02600195).18 All hospitals were national or provincial neonatal referral centers or regional referral centers in metropolitan cities. All hospitals provided comprehensive care for infants with gestational age <28 weeks or birth weight <1000 g and had advanced respiratory support including high-frequency ventilation. This study was a retrospective evaluation of this prospectively collected data. The study was approved by the Ethics Committee of the Children’s Hospital of Fudan University and recognized by all participating centers.
The inclusion criteria were all infants born at <34 weeks of gestation between May 1, 2015, and April 30, 2018, and admitted to participating NICUs within 7 days of birth. Infants were excluded if they were discharged against medical advice (received incomplete care in the NICU), stayed in the NICU for fewer than 3 days or were transferred to a nonparticipating NICU with outcome data unavailable.
Infants with early-onset culture-positive sepsis, definite local infections or infection-related morbidities were also excluded to avoid confounding the outcome measures. Local infections or infection-related morbidities included ventilator-associated pneumonia diagnosed based on modified criteria from the US Centers for Disease Control and Prevention19 and the published literature,20 urinary tract infection (defined as bacterial counts >104 colony-forming units/mL from urine obtained by sterile catheter) and stage II or III necrotizing enterocolitis. Finally, infants with more than 1 episode of LOS (either culture-negative or culture-positive LOS) during their hospitalizations were excluded to allow comparison of the influence of a single episode of culture-negative or culture-positive LOS on the neonatal outcomes.
Infants were followed until death or discharge from the NICU.
Detailed clinical data of all enrolled preterm infants were collected prospectively by trained data abstractors using a standardized database. All data collection followed standard operations and definitions.
For all participating NICUs, an algorithm for workup and diagnosis of LOS was introduced and recommended at the beginning of the study (Figure, Supplemental Digital Content 1, http://links.lww.com/INF/D727). The algorithm included a list of clinical manifestations that might indicate infection. Laboratory tests recommended when LOS was suspected included whole blood cell count, C-reactive protein (CRP), procalcitonin (PCT), blood and urine culture, lumbar puncture and chest and abdominal radiograph. One blood culture bottle with at least 1 mL of blood was recommended for each blood culture.
LOS was defined as sepsis occurring after 3 days of age. Culture-negative sepsis was diagnosed when all the following criteria were fulfilled: (1) 2 or more infection-related clinical manifestations; (2) abnormal white blood cell count, CRP level or PCT level; (3) antibiotics used or intended for ≥5 days; (4) negative blood culture with no or negative cerebrospinal fluid culture; (5) no evidence of concurrent focal infection, including pneumonia, urinary tract infection and necrotizing enterocolitis. Culture-positive sepsis was defined by a positive blood or cerebrospinal fluid culture. Cultures that were positive for possible contaminants, mainly coagulase-negative staphylococci, were considered indicative of infection if the infants fulfilled all 3 of the following criteria: (1) 1 or more infection-related clinical manifestations; (2) 2 positive cultures drawn within 2 days of each other, or 1 positive blood culture together with an abnormal white blood cell count, CRP level or PCT level; (3) susceptible antibiotics received for ≥5 days.21 Date of LOS onset was defined as the date of blood culture. An antibiotic course consisted of 1 or more uninterrupted antibiotic days. The Transport Risk Index of Physiologic Stability (TRIPS) score22 was used as an illness severity score on NICU admission.
The primary outcome was a composite of mortality or 1 of 3 major morbidities: periventricular leukomalacia (PVL), ROP ≥ stage 3 or BPD. We defined PVL as the presence of periventricular cysts on cranial ultrasound or cranial magnetic resonance imaging scans before discharge. We defined ROP according to the International Classification of ROP.23 We defined BPD as mechanical ventilation or oxygen dependency at 36 weeks’ postmenstrual age or discharge.24 These 3 morbidities were chosen, because they were associated with systemic inflammation and usually occurred after the median age of infection in preterm infants. The secondary outcomes were rates of mortality and the individual morbidities, and NICU resource use, including NICU stay and antibiotic use.
Infant and maternal characteristics and outcomes were compared among infants with culture-negative LOS, culture-positive LOS and no LOS using the χ2 test for categorical variables and the Kruskal–Wallis test for continuous variables. Group pair-wise comparisons were further conducted where the omnibus test was significant. For the purpose of multiple comparison adjustment, a P-value of 0.017 was considered significant according to the Bonferroni method. Multilevel mixed-effects logistic regression models were used to examine the association of culture-negative LOS and neonatal outcomes accounting for the intracluster correlation among the infants within hospitals. Hospitals were considered as independent clusters with random effects in the models. At the infant level, we controlled for sex, gestational age, small for gestational age, Apgar score <3 at 5 minutes, TRIPS score on admission, inborn, cesarean, maternal hypertension, maternal diabetes and antenatal steroids. A 2-sided P-value of <0.05 was used to determine statistical significance. Statistical analyses were performed using Stata 13.1 (StataCorp, 2013; College Station, TX, USA).
A total of 27,532 infants with gestational age <34 weeks were assessed for eligibility, and 5186 infants were excluded (Fig. 1). The remaining 22,346 infants composed the study population, among which 1505 (6.7%) infants had 1 episode of culture-negative LOS, 761 (3.4%) infants had 1 episode of culture-positive LOS and 20,080 (89.9%) infants did not have any episode of LOS during their hospitalization (Fig. 1). The incidences of culture-negative LOS were higher than those of culture-positive LOS among infants in different birth weight categories (Fig. 2).
Infant and maternal characteristics are shown in Table 1. Infants with culture-negative or culture-positive LOS were of significantly lower gestational age and lower birth weight, more likely to be small for their gestational age and to be outborn, and had higher TRIPS score on admission compared with infants without LOS. There were no significant differences in baseline characteristics between infants with culture-negative LOS and culture-positive LOS, except the male and inborn rates.
Compared with infants without LOS, those with culture-negative LOS had higher rates of composite outcome, death, PVL, ROP ≥ stage 3 and BPD (Table 2). Compared with infants with culture-positive LOS, those with culture-negative LOS had lower rates of composite outcome, death and BPD, and similar rates of PVL and ROP ≥ stage 3. The median duration of antibiotic course for each episode of culture-negative LOS (11 days; interquartile range 8–16 days) was shorter than that for episodes of culture-positive LOS (16 days; interquartile range 11–22 days, P < 0.0001) (Table 2). Among infants with culture-negative LOS, 17.9% (270/1505) and 9.2% (139/1505) required mechanical ventilation and inotropes, respectively, within 3 days of diagnosis, lower than for infants with culture-positive LOS (Table 2).
Table 3 shows the results of multilevel mixed-effects logistic regression analysis. Compared with the absence of LOS, culture-negative LOS was independently associated with increased odds of composite outcome, PVL and BPD in the adjusted analysis. Infants with culture-positive LOS also showed higher adjusted odds of composite outcome, death and BPD than those without LOS. Compared with culture-negative LOS, infants with culture-positive LOS were at higher risk of death, while the risks of composite outcome and the individual morbidities were not statistically significantly different between these 2 groups.
Our study investigated the contemporary incidence and outcomes of culture-negative LOS in a large cohort of preterm infants. We identified that culture-negative LOS was diagnosed approximately twice as frequently as culture-positive LOS in preterm infants. Culture-negative LOS was associated with increased odds of adverse outcomes compared with the absence of LOS.
One major difficulty in reporting culture-negative LOS is the lack of consensus on diagnostic criteria. Some studies, which used antibiotic use as the definition of culture-negative LOS, may have overestimated its occurrence.11,14 One large trial diagnosed culture-negative LOS based on use of antibiotics and clinical manifestations,25 which is highly reliant on the judgments of physicians. Although physicians’ abilities to identify sepsis in neonates is suboptimal at the onset of infection, the accuracy of clinical judgment increased significantly over time.26 More recent studies defined culture-negative LOS using the combined criteria of clinical signs, abnormal laboratory results and antibiotic treatment10,12,27,28 and suggested they be used in neonatal sepsis trials.29 During our study period, a standardized algorithm of LOS workup was suggested to all participating NICUs. The diagnosis of culture-negative LOS was based on all 3 criteria of clinical signs, laboratory parameters and use of antibiotics. White blood cell count, CRP and PCT were widely used in almost all participating NICUs and therefore were chosen as the required laboratory tests. However, we must acknowledge that although we used all currently existing and clinically available parameters in participating NICUs, there was still a risk of misclassifying uninfected infants to the culture-negative LOS group.
We observed the incidence of culture-negative sepsis to be 6.7% among infants <34 weeks of gestation age and as high as 17.4% for infants with birth weight <1000 g. As we excluded infants with early-onset sepsis, local infections, necrotizing enterocolitis and multiple episodes of LOS in our study population, our reported incidence of culture-negative LOS is likely to be still an underestimation. Additionally, our results showed culture-negative LOS was associated with increased risks of composite adverse outcome, PVL and BPD compared with infants without LOS. These findings suggest that culture-negative LOS could be one or a composite of true pathologic conditions associated with adverse outcomes in preterm infants and should not be simply ignored in studies regarding neonatal sepsis.
However, the unanswered questions are: What actually causes culture-negative LOS? And if it is a true infection, why are the culture results negative? Although our study cannot definitively answer these questions, the findings provide insight into some of the possibilities.
First, it is possible that cultures may be falsely negative because of inappropriate sampling practices such as obtaining insufficient blood volume and performing cultures after antibiotics have been started.1,2,13 This may result in both a low incidence of culture-positive LOS and a correspondingly high incidence of culture-negative LOS. Therefore, quality improvement efforts to facilitate appropriate sampling practices may need to be prioritized when managing the problem of culture-negative LOS.
Second, culture-negative sepsis may reflect a milder form of LOS compared with culture-positive LOS.30 Our results showed that, compared with infants with culture-positive LOS, those with culture-negative LOS were at lower risk of death and required less intensive support such as invasive ventilation and inotropes. Similar findings were reported in adults: patients with culture-negative sepsis had fewer comorbidities and lower severity of illness than those with culture-positive sepsis.30 Misclassification of culture-negative and culture-positive sepsis could occur because of sampling errors; however, these 2 groups are not identical. Bacteremia with low colony counts accounted for ~70% of bacteremia in infants younger than 2 months.31 A recent study showed that 16S rRNA quantitative polymerase chain reaction was able to identify pathogens in 85% of neonatal culture-negative sepsis, but the bacterial loads were significantly lower in culture-negative sepsis compared with culture-positive sepsis.32 Negative neonatal blood cultures can also be because of transiently or intermittently positive cultures. Therefore, the lower severity of culture-negative LOS might be partly because of a lower bacterial burden and a milder insult corresponding to the negative culture results, especially in this vulnerable preterm population with generally compromised immune function and increased susceptibility to invasive infections. This again emphasizes the importance of observing appropriate culture practices to maximize the sensitivity of blood cultures. Other new pathogen-detecting technologies, such as real-time polymerase chain reaction and next-generation sequencing, may also have roles in the future to improve the pathogen diagnosis of culture-negative LOS.
Third, some infants diagnosed with culture-negative LOS may have had nonbacterial infections such as viral infections or infections that are not readily cultured using conventional blood culture methods. However, although viral infections were reported to occur in about 1% of NICU admissions,33 the majority of these infections were community acquired and caused local infection such as respiratory syncytial virus-related pneumonitis.33 Although we cannot rule out the possibility of viral infection presenting as culture-negative LOS because viral infection data were not collected, all hospitals in our study had the capacity to detect infectious viruses.
Culture-negative LOS contributes significantly to high rates of antibiotic use in NICUs,14,17,34 and controversy remains about whether infants with culture-negative LOS need antibiotic treatment.13 Although our results showed that culture-negative LOS was associated with adverse outcomes, given the heterogeneous conditions underlining culture-negative LOS, we could not answer the question definitely whether it is reasonable to treat culture-negative LOS with antibiotics. However, it does not mean that there is nothing we could do to facilitate rational antibiotic use in these infants. A recent study on antimicrobial stewardship in NICUs suggested that 5 days of antibiotics should be enough for treatment of culture-negative LOS.17 However, in our study, the median duration of antibiotic treatment for each episode of culture-negative LOS was 11 days. This signifies that we urgently need to initiate antibiotic stewardship measures in our NICUs for culture-negative LOS to at least shorten the inappropriate duration of antibiotics.
Strengths of our study include that we predefined culture-negative LOS and collected the information on this measure prospectively and that we compared outcomes of infants with culture-negative LOS with outcomes of those without LOS and those with culture-positive LOS. However, we acknowledge the limitations of our study. First, although a relatively strict definition of culture-negative LOS was used, there may have been some misclassifications of infants with or without true infection. Nonetheless, this mirrors the real-world phenomenon. Second, we did not standardize the performance of blood cultures among participating hospitals. Third, we did not collect data on viral infection. Fourth, NICUs in China tended to use relatively more with longer duration of antibiotics than NICUs in developed countries.17 This may influence the generalization of our results to NICUs using fewer antibiotics.
In conclusion, our study confirmed that culture-negative LOS was frequently diagnosed and treated with long durations of antibiotics in preterm infants and was associated with adverse neonatal outcomes. Therefore, effectively managing this condition will require attention in several areas. In particular, quality improvement processes are needed to facilitate optimal blood culture practices, to help reach consensus on appropriate diagnostic criteria and to optimize antibiotic treatment for culture-negative LOS.
In addition to the listed authors, the REIN-EPIQ Study Group includes the following investigators: Yong Ji, MD, Children’s Hospital of ShanXi / Women’s Health Center of Shanxi, Shanxi, China; Shuping Han, MD, Women’s Hospital of Nanjing Medical University, Jiangsu, China; Sannan Wang, MD, Suzhou Municipal Hospital, Jiangsu, China; Zhankui Li, MD, Northwest Women and Children’s Hospital, Shaanxi, China; Shiwen Xia, MD, Women and Children’s Hospital of Hubei Province, Hubei, China; Changyi Yang, MD, Fujian Provincial Maternity and Children’s Hospital/Affiliated Hospital of Fujian Medical University, Fujian, China; Chuanzhong Yang, MD, The Affiliated Shenzhen Maternity and Child Healthcare Hospital of Southern Medical University, Guangdong, China; Ling Chen, MD, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Hubei, China; Jing Yuan, MD, Qingdao Women and Children’s Hospital, Shandong, China; Ling Liu, MD, Guiyang Maternal and Child Health Care Hospital, Guizhou, China; Bin Yi, MD, Gansu Provincial Maternity and Child-care Hospital, Gansu, China; Zhenlang Lin, MD, The 2nd Affiliated Hospital and Yuying Children’s Hospital of Wenzhou Medical University, Zhejiang, China; Yang Wang, MD, The First Affiliated Hospital of Anhui Medical University, Anhui, China; Jiangqin Liu, MD, Shanghai First Maternity and Infant Hospital, Tongji University School of Medicine, Shanghai, China; Ling He, MD, Jiangxi Provincial Children’s Hospital, Jiangxi, China; Mingxia Li, MD, First Affiliated Hospital of Xinjiang Medical University, Xinjiang, China; Xinnian Pan, MD, The Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Guangxi, China; Yan Guo, MD, Children’s Hospital of Nanjing Medical University, Jiangsu, China; Cuiqing Liu, MD, Children’s Hospital of Hebei Province, Hebei, China; Qin Zhou, MD, The Affiliated Wuxi Maternity and Child Health Hospital of Nanjing Medical University, Jiangsu, China; Xiaoying Li, MD, Qilu Children’s Hospital of Shandong University, Shandong, China; Hong Xiong, MD, Children’s Hospital Affiliated to Zhengzhou University, Henan, China; Yujie Qi, MD, Beijing Children’s Hospital of Capital Medical University, Beijing, China; Mingyan Hei, MD, The Third Xiangya Hospital of Central South University, Hunan, China. The authors gratefully acknowledge the Canadian Neonatal Network for their support of the REIN-EPIQ study and all the data abstractors of the REIN-EPIQ study. In addition, we thank Heather McDonald Kinkaid, PhD, from the Maternal-Infant Care Research Centre at Mount Sinai Hospital, Toronto, Ontario, Canada, for editorial assistance in the preparation of this manuscript.
1. Connell TG, Rele M, Cowley D, et al. How reliable is a negative blood culture result? Volume of blood submitted for culture in routine practice in a children’s hospital. Pediatrics. 2007;119:891–896.
2. Schelonka RL, Chai MK, Yoder BA, et al. Volume of blood required to detect common neonatal pathogens. J Pediatr. 1996;129:275–278.
3. Wynn JL, Polin RA. Progress in the management of neonatal sepsis: the importance of a consensus definition. Pediatr Res. 2018;83:13–15.
4. Klingenberg C, Kornelisse RF, Buonocore G, et al. Culture-negative early-onset neonatal sepsis - at the crossroad between efficient sepsis care and antimicrobial stewardship. Front Pediatr. 2018;6:285.
5. Stocker M, van Herk W, El Helou S, et al; NeoPInS Study Group. Procalcitonin-guided decision making for duration of antibiotic therapy in neonates with suspected early-onset sepsis: a multicentre, randomised controlled trial (NeoPIns). Lancet. 2017;390:871–881.
6. Stoll BJ, Hansen N, Fanaroff AA, et al. Late-onset sepsis
in very low birth weight neonates: the experience of the NICHD neonatal research network. Pediatrics. 2002;110(2 pt 1):285–291.
7. Hornik CP, Fort P, Clark RH, et al. Early and late onset sepsis in very-low-birth-weight infants from a large group of neonatal intensive care units. Early Hum Dev. 2012;88(suppl 2):S69–S74.
8. Giannoni E, Agyeman PKA, Stocker M, et al; Swiss Pediatric Sepsis Study. Neonatal sepsis of early onset, and hospital-acquired and community-acquired late onset: a Prospective Population-Based Cohort Study. J Pediatr. 2018;201:106–114.e4.
9. Stoll BJ, Hansen NI, Bell EF, et al; Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network. Trends in care practices, morbidity, and mortality of extremely preterm neonates, 1993-2012. JAMA. 2015;314:1039–1051.
10. Griffiths J, Jenkins P, Vargova M, et al. Enteral lactoferrin supplementation for very preterm infants: a randomised placebo-controlled trial. Lancet. 2019;393:423–433.
11. Stoll BJ, Hansen NI, Adams-Chapman I, et al; National Institute of Child Health and Human Development Neonatal Research Network. Neurodevelopmental and growth impairment among extremely low-birth-weight infants with neonatal infection. JAMA. 2004;292:2357–2365.
12. Schlapbach LJ, Aebischer M, Adams M, et al; Swiss Neonatal Network and Follow-Up Group. Impact of sepsis on neurodevelopmental outcome in a Swiss National Cohort of extremely premature infants. Pediatrics. 2011;128:e348–e357.
13. Cantey JB, Baird SD. Ending the culture of culture-negative sepsis
in the neonatal ICU. Pediatrics. 2017;140:e20170044.
14. Wirtschafter DD, Padilla G, Suh O, et al. Antibiotic use for presumed neonatally acquired infections far exceeds that for central line-associated blood stream infections: an exploratory critique. J Perinatol. 2011;31:514–518.
15. Ting JY, Synnes A, Roberts A, et al; Canadian Neonatal Network Investigators. Association between antibiotic use and neonatal mortality and morbidities in very low-birth-weight infants without culture-proven sepsis or necrotizing enterocolitis. JAMA Pediatr. 2016;170:1181–1187.
16. Ting JY, Roberts A, Sherlock R, et al. Duration of initial empirical antibiotic therapy and outcomes in very low birth weight infants. Pediatrics. 2019;143:e20182286.
17. Cantey JB, Wozniak PS, Pruszynski JE, et al. Reducing unnecessary antibiotic use in the neonatal intensive care unit (SCOUT): a prospective interrupted time-series study. Lancet Infect Dis. 2016;16:1178–1184.
18. REIN-EPIQ Study Group. Reduction of infection in neonatal intensive care units using the evidence-based practice for improving quality (REIN-EPIQ): a study protocol. Chin J Evid Based Pediatr. 2018;13:452–457.
19. Magill SS, Klompas M, Balk R, et al. Developing a new national approach to surveillance for ventilator-associated events: executive summary. Am J Infect Control. 2013;41:1096–1099.
20. Zhou Q, Lee SK, Jiang SY, et al. Efficacy of an infection control program in reducing ventilator-associated pneumonia in a Chinese neonatal intensive care unit. Am J Infect Control. 2013;41:1059–1064.
21. Stoll BJ, Hansen N, Fanaroff AA, et al. Changes in pathogens causing early-onset sepsis in very-low-birth-weight infants. N Engl J Med. 2002;347:240–247.
22. Lee SK, Aziz K, Dunn M, et al; Canadian Neonatal Network. Transport Risk Index of Physiologic Stability, version II (TRIPS-II): a simple and practical neonatal illness severity score. Am J Perinatol. 2013;30:395–400.
23. Gole GA, Ells AL, Katz X, et al. The international classification of retinopathy of prematurity revisited. Arch Ophthalmol. 2005;123:991–999.
24. J Jobe AH, Bancalari E. Bronchopulmonary dysplasia. Am J Respir Crit Care Med. 2001;163:1723–1729.
25. Brocklehurst P, Farrell B, King A, et al; The INIS Collaborative Group. Treatment of neonatal sepsis with intravenous immune globulin. N Eng J Med. 2011;365:1201–1211.
26. Fischer JE. Physicians’ ability to diagnose sepsis in newborns and critically ill children. Pediatr Crit Care Med. 2005;6(3 suppl):S120–S125.
27. Rueda MS, Calderon-Anyosa R, Gonzales J, et al; NEOLACTO Research Group. Antibiotic overuse in premature low birth weight infants in a developing country. Pediatr Infect Dis J. 2019;38:302–307.
28. Bekhof J, Reitsma JB, Kok JH, et al. Clinical signs to identify late-onset sepsis
in preterm infants. Eur J Pediatr. 2013;172:501–508.
29. Oeser C, Lutsar I, Metsvaht T, et al. Clinical trials in neonatal sepsis. J Antimicrob Chemother. 2013;68:2733–2745.
30. Phua J, Ngerng W, See K, et al. Characteristics and outcomes of culture-negative versus culture-positive severe sepsis. Crit Care. 2013;17:R202.
31. Kellogg JA, Ferrentino FL, Goodstein MH, et al. Frequency of low level bacteremia in infants from birth to two months of age. Pediatr Infect Dis J. 1997;16:381–385.
32. Stranieri I, Kanunfre KA, Rodrigues JC, et al. Assessment and comparison of bacterial load levels determined by quantitative amplifications in blood culture-positive and negative neonatal sepsis. Rev Inst Med Trop Sao Paulo. 2018;60:e61.
33. Verboon-Maciolek MA, Krediet TG, Gerards LJ, et al. Clinical and epidemiologic characteristics of viral infections in a neonatal intensive care unit during a 12-year period. Pediatr Infect Dis J. 2005;24:901–904.
34. Różańska A, Wójkowska-Mach J, Adamski P, et al. Antibiotic consumption in laboratory confirmed vs. non-confirmed bloodstream infections among very low birth weight neonates in Poland. Ann Clin Microbiol Antimicrob. 2017;16:20.