Original Article
675
Risk Factors and Outcomes of Late-Onset Bacterial Sepsis in Preterm Neonates Born at < 32 Weeks’ Gestation Jyotsna Shah, MD, CCRP, DTMH1 Ann L. Jefferies, MD, MEd, FRCPC1 Eugene W. Yoon, MSc2 Shoo K. Lee, MBBS, PhD, FRCPC1,2 Prakesh S. Shah, MD, MSc, MRCP, FRCPC1,2 and on behalf of the Canadian Neonatal Network
Toronto, Toronto, Canada 2 Maternal-Infant Care Research Centre, Mount Sinai Hospital, Toronto, Canada
Address for correspondence Jyotsna Shah, MD, CCRP, DTMH, Department of Pediatrics, Mount Sinai Hospital, Rm 19-231, 600 University Avenue, Toronto, ON M5G 1X5, Canada (e-mail: [email protected]).
Am J Perinatol 2015;32:675–682.
Abstract
Keywords
► ► ► ► ► ►
infant premature nosocomial infection late-onset sepsis mortality bronchopulmonary dysplasia
Objective This study aims to identify the incidence, risk factors, and outcomes of lateonset sepsis in preterm neonates in Canadian neonatal intensive care units (NICUs). Study Design This retrospective analysis included preterm infants born at < 32 weeks’ gestation and admitted to 29 NICUs in the Canadian Neonatal Network during the years 2010 and 2011. Infants were classified into three groups: no infection, grampositive infection, and gram-negative infection. Late-onset sepsis was defined as positive blood and/or spinal fluid cultures after 3 days of birth. Risk factors and the primary outcome of mortality or bronchopulmonary dysplasia (BPD) were compared between the groups. Results Out of the 7,509 neonates, 6,405 (85%) had no infection, 909 (12%) had grampositive, and 195 (3%) had gram-negative infections. Lower gestation, higher Score for Neonatal Acute Physiology, version II scores, the presence of central catheters for > 4 days, parenteral nutrition for > 7 days, and prolonged duration of nothing by mouth were associated with late-onset sepsis. After controlling for confounders, the odds ratio (OR) of mortality/BPD were higher in infants who had gram-negative (OR 2.79, 95% confidence interval [CI] 1.96–3.97) and gram-positive (OR 1.44, 95% CI 1.21–1.71) sepsis as compared with no infection. Conclusions Bacterial late-onset sepsis in very preterm neonates was associated with mortality and BPD. Neonates with gram-negative sepsis had the highest risk of adverse outcomes as compared with gram-positive sepsis or no sepsis.
Late-onset sepsis (LOS) is a major cause of mortality and morbidity in the neonatal intensive care unit (NICU). The incidence of LOS is reported to vary from 20 to 50%, depending upon the patient population, with higher rates among lower gestation and lower birth weight infants.1–6 Data from the Neonatal Network of the National Institute of Child Health
and Human Development in the United States reported that LOS occurred in almost 25% of very low-birth-weight infants.7 The epidemiology of the pathogens responsible for LOS has changed in the past 75 years.8 In the early 1950s, the organisms most responsible for neonatal sepsis were Staphylococcus sp., in the 1960s it was gram-negative
received July 10, 2014 accepted after revision August 28, 2014 published online December 8, 2014
Copyright © 2015 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI http://dx.doi.org/ 10.1055/s-0034-1393936. ISSN 0735-1631.
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
1 Department of Pediatrics, Mount Sinai Hospital, University of
Late-Onset Bacterial Sepsis in Preterm Neonates
Shah et al.
organisms, and since the 1970s coagulase-negative Staphylococcus sp. have emerged as the most common sepsis-causing organisms. In recent years, it has been reported that among cases of LOS in very low-birth-weight infants, 20 to 30% are due to gram-negative organisms, whereas as much as 50% of early-onset infections are due to gram-positive organisms.9–11 No increase in the overall rates of early-onset sepsis or LOS has been reported in recent years subsequent to an increased rate of antepartum or intrapartum chemoprophylaxis for group B streptococcus; however, it is suspected that this may have led to the emergence of resistant strains of gram-negative bacteria.1,12 We have previously identified a steady decline in the rate of LOS in Canadian NICUs (from 4.5 infections/1,000 days in 2003 to 3.5/1,000 days in 2009, a 20% reduction)13; however, it is important to monitor and conduct regular surveillance of the outcomes of infants who develop LOS compared with those without sepsis.14 The risk factors for LOS in preterm neonates include extreme prematurity, low birth weight, use of invasive procedures, including intubation, the presence of central lines, need for total parental nutrition with lipids, receipt of broadspectrum antibiotics, and delayed initiation of enteral feed.1,12,15–17 In addition, the morbidity and mortality of neonates with gram-negative sepsis is high. The invasive and fulminant course of certain gram-negative infections in preterm infants is often devastating.18 Recently, Levit et al reported that need for intubation, pressors, hypoglycemia, and thrombocytopenia were associated with mortality among LOS neonates.11 According to a report from the National Institute of Child Health and Human Development (NICHD) network NICUs between 1998 and 2000, there were higher rates of intraventricular hemorrhage, patent ductus arteriosus, necrotizing enterocolitis (NEC), and duration of hospitalization, especially those who had gram-negative sepsis.10 Contemporary information regarding the rates, risk factors, and outcomes of neonates with LOS caused by bacteria, especially gram-negative sepsis, is needed to counsel parents and to develop indicators for further quality improvement, as well as preventative strategies.14 Our objective of this study was to identify the associated risk factors and outcomes of preterm infants born at < 32 weeks’ gestation who developed LOS during their stay in NICUs in Canada.
Patients and Methods Population We conducted a retrospective analysis of preterm neonates born at < 32 weeks’ gestation and admitted to 29 level-3 NICUs participating in the Canadian Neonatal Network (CNN) during the years 2010 and 2011 (2-year period). We excluded infants who were moribund (palliative care planned at birth), infants with major congenital anomalies, and infants who had early-onset sepsis (blood or spinal fluid culture-positive within the first 3 days of birth).
Exposure (Infecting Pathogen) Infants were identified as having LOS when either blood and/ or cerebrospinal fluid cultures were positive for a pathogenic American Journal of Perinatology
Vol. 32
No. 7/2015
organism in a symptomatic infant after 3 days of birth. The cohort was divided into three groups: neonates who had no infections during their stay in a participating NICU, neonates with gram-positive infections, and neonates with gram-negative infections. Infants who had only viral, mycoplasma, ureaplasma, or fungal infections were excluded as the objective of this study was to assess the impact of bacterial infections. Infants who were infected with both gram-positive and gram-negative organisms were assigned to the gramnegative group. The presence of an organism in one blood or spinal fluid culture was adequate to diagnose LOS. Due to the small size of infants, the technical challenges associated with drawing samples for two blood cultures and the fact that a single positive culture is considered adequate for the diagnosis of neonatal sepsis,19 a second positive culture was not mandatory.
Risk Factor Evaluation To assess the risk factors for LOS, the following information was obtained from the CNN database: duration of central catheter (dichotomized at < or 4 days), duration of nothing by mouth, severity of illness using the Score for Acute Neonatal Physiology, version II (SNAPII) score,20 duration of parenteral nutrition (dichotomized at < 7 or 7 days), maternal chorioamnionitis (defined as maternal fever during peripartum period, and either foul smelling liquor, uterine tenderness or high maternal white cell count), maternal antibiotics during labor, and small for gestational age, defined as birth weight below the 10th centile for gestational age and sex. These risk factors and outcomes were compared between the three groups. Dichotomization for the central catheter duration and duration of parenteral nutrition was based on median duration of the respective variables in the network data.
Outcomes The primary outcome of this study was a composite of mortality and bronchopulmonary dysplasia (BPD). Mortality was defined as any neonate who died 3 days or more after birth, but before discharge from the NICU. BPD was defined as supplemental oxygen dependency at 36 weeks corrected age or at the time of transfer to another medical facility if that occurred before 36 weeks’ corrected gestational age.21 The secondary outcomes were BPD, severe (> stage 2) retinopathy of prematurity (ROP), stage 2 or 3 NEC, and the following measures of resource utilization: duration of mechanical ventilation, duration of respiratory support, and length of stay. Gestational age was defined as the best estimate based on obstetric history, obstetric examination, and first prenatal ultrasound examination. Severity of illness was assessed using the SNAPII score. ROP was defined according to the international classification.22 NEC was defined according to the Bell criteria.23 Duration of mechanical ventilation was defined as the total number of days during which the infant was on mechanical ventilation (not including noninvasive modes) during any part of the day. Duration of respiratory support was defined as the duration of invasive respiratory
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
676
Late-Onset Bacterial Sepsis in Preterm Neonates
Shah et al.
677
support, continuous positive pressure airway pressure support or other noninvasive respiratory support. Length of stay was defined as the total number of days that an infant stayed in the NICU before death, discharge to home, or transfer to a level-2 hospital.
Data on individual infants were collected as part of the CNN’s ongoing data collection system. Out of the 30 NICUs in Canada, 29 NICUs were participating in the network during the study period. At all affiliated sites, demographic and outcomes data were collected from patient charts by trained research assistants using a computerized data entry program according to the standardized outcome definitions.24 Data were transmitted electronically to the CNN coordinating center where they were stored. Data were collected on each infant until death or discharge from the NICU. Data collection was approved by each institution’s research ethics board or institutional quality improvement committee as appropriate.
Analyses Statistical analyses were performed with SAS software 9.2 (SAS Institute, Inc., Cary, NC). Baseline demographic characteristics, perinatal risk factors, and the incidence of primary and secondary outcomes among infants in the no infection (reference group), gram-positive infection, and gram-negative infection groups were compared using Pearson chisquare test for categorical variables and Student’s t-test or analysis of variance for continuous variables. Multivariate logistic regression analysis was used to examine the effect of clinically significant and other confounders identified from univariate analyses, which included gestational age, sex, SNAPII 20, small for gestational age, central catheter duration > 4 days, and parenteral nutrition > 7 days.
Results Out of the 33,633 neonates admitted to NICUs in Canada during the study time period, 7,509 neonates were included in the study cohort as described in ►Fig. 1. Of these, 6,405 (85%) neonates had no infections, 909 (12%) had gram-positive infections, and 195 (3%) had gram-negative infections. Of the 1,114 neonates with bacterial LOS, 1,034 (94%) were blood-culture positive, 21 (2%) were both blood-culture and spinal-fluid positive, and 49 (4%) were only spinal-fluid positive for organisms. Of the 1,034 neonates with positive blood culture 583 had spinal fluid evaluated. Comparison of baseline characteristics between the three groups revealed that infants who had infections had a lower gestational age, lower birth weight, and higher severity of illness at admission (SNAPII score > 20). In addition, assessment of risk factors revealed that a higher number of infants with infection had longer durations of central catheter use, parenteral nutrition administration, and nothing by mouth, as well as higher rates of maternal chorioamnionitis (►Table 1). Comparison of the primary outcomes of mortality/BPD and the secondary outcomes revealed that the outcome rates
Fig. 1 Patient selection.
were significantly higher in the infection groups with the worst outcomes occurring in infants with gram-negative infections (►Table 2). Resource utilization data also revealed significantly higher durations of respiratory support and length of stay for infants with infection, especially those with gram-negative infections (►Table 2). After adjusting for confounders and baseline differences, there were significantly higher odds of mortality/BPD among infants with both gram-positive and gram-negative infections as compared with no infection (►Table 3). Comparison of patients with gram-positive infections to those with gramnegative infections revealed that the combined outcome of mortality/BPD, and BPD and mortality individually were significantly higher in infants with gram-negative infections (►Table 3). Organisms and their frequency are reported in ►Table 4. The organisms found most frequently in infants in the grampositive group were coagulase-negative Staphylococci, coagulase-positive Staphylococcus aureus, and Enterococci, whereas in the gram-negative group Escherichia coli, Klebsiella, and Enterobacter were found most frequently (►Table 4).
Discussion In this large population-based cohort of very preterm neonates, we identified that LOS still remains a major cause of mortality and morbidity. Among the infants with bacterial LOS, 80% were caused by gram-positive organisms, whereas 20% were caused by gram-negative organisms. Analysis of the outcomes of infants who had bacterial LOS revealed higher odds of mortality, BPD, severe ROP, and NEC, as well as higher resource utilization compared with infants who did not have sepsis. Lower gestational age, higher SNAPII scores, the presence of central catheters for > 4 days (median duration of the cohort), parenteral nutrition administration for > 7 days (median duration of the cohort), and prolonged duration of nothing by mouth were significant risk factors associated with bacterial LOS. The most common gram-positive American Journal of Perinatology
Vol. 32
No. 7/2015
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
Data Collection
Late-Onset Bacterial Sepsis in Preterm Neonates
Shah et al.
Table 1 Maternal and neonatal characteristics of study infants No infection (N ¼ 6,405)
GP infection (N ¼ 909)
GN infection (N ¼ 195)
p-Value None vs. GP
p-Value None vs. GN
p-Value GP vs. GN
Antenatal steroids, n (%)
5,378 (87)
768 (88)
167 (87)
0.9741
0.8761
0.8745
Multiples, n (%)
1,901 (30)
273 (30)
63 (32)
0.9227
0.4089
0.4762
Maternal chorioamnionitis, n (%)
889 (19)
137 (20)
43 (28)
0.2719
0.0044
0.0465
Vaginal births, n (%)
2,630 (41)
353 (39)
104 (53)
0.1485
0.0005
0.0001
Gestational age (wk), mean (SD)
28.5 (2.2)
26.8 (2.2)
26.4 (2.2)
< 0.0001
< 0.0001
0.0167
Birth weight (g), mean (SD)
1,240 (394)
978 (314)
956 (337)
< 0.0001
< 0.0001
0.4446
Female, n (%)
2,925 (46)
392 (43)
85 (44)
0.1490
0.3704
0.8623
Small for gestational age, n (%)
593 (9)
116 (13)
19 (10)
0.0006
0.9358
0.1401
Apgar score < 7 at 5 min, n (%)
1,716 (27)
349 (39)
71 (37)
< 0.0001
0.0016
0.7201
SNAP score, median (interquartile range)
9 (0, 14)
14 (7, 21)
14 (9, 24)
< 0.0001
< 0.0001
0.1309
SNAPII score 20, n (%)
1,013 (16)
251 (28)
68 (35)
< 0.0001
< 0.0001
0.0559
Duration of central catheter, median (interquartile range)
7 (0, 14)
22 (10, 38)
24 (12, 48)
< 0.0001
< 0.0001
0.2520
Duration of central catheter > 4 d, n (%)
3,735 (58)
789 (87)
180 (92)
< 0.0001
< 0.0001
0.0517
Duration of nothing by mouth during stay, median (interquartile range)
1 (1, 3)
6 (2, 12)
8 (3, 16)
< 0.0001
< 0.0001
0.1161
Duration of parenteral nutrition, median (interquartile range)
10 (6, 16)
28 (19, 45)
30 (15, 53)
< 0.0001
< 0.0001
0.8798
Duration of parenteral nutrition > 7 d, n (%)
4,062 (63)
876 (96)
176 (90)
< 0.0001
< 0.0001
0.0003
Characteristics Maternal/peripartum variables
Birth variables
Neonatal course
Abbreviations: GN, gram-negative; GP, gram-positive; SNAPII, Score for Acute Neonatal Physiology version II.
organism was coagulase-negative Staphylococci and the most common gram-negative organism was E. coli. Several single center and multicenter reports, from developed countries of rates of LOS and the contribution of grampositive and gram-negative organisms, are available with varying estimates of the incidence of LOS that range from 25 to 35% in very low-birth-weight infants.4,5,25 In these reports, gram-positive organisms accounted for 55 to 65% of the cases, gram-negative organisms for 25 to 35% of the cases, and fungi, mycoplasma, ureaplasma, and, rarely, viral organisms for the remaining cases.3,4,26 Rates of gram-negative organisms have been reported to be as high as 46%.27 Among the gram-positive organisms, coagulase-negative Staphylococci has been reported to be responsible for > 50% of LOS in preterm infants5; however, the distribution of gramnegative organisms is variable. Pseudomonas,5,6 Klebsiella,4,27 E. coli, and Enterobacter species17,28 have been reported to be the predominant organisms from various units. In our cohort, American Journal of Perinatology
Vol. 32
No. 7/2015
among the infants with bacterial LOS, gram-negative organisms accounted for 20% of the cases, slightly lower than previous reports from other networks and single centers. We speculate that this reduction is a result of our networkwide quality improvement efforts over 2009 to 2012 and suggests a continued decline in rates of LOS.13 However, there was no statistically significant difference in the rate of infection between two years of this study period. Our rate of coagulase-negative Staphylococci was higher than reported in other studies, which could be because we have not mandated two cultures in the diagnosis of sepsis. Ongoing surveillance like this report is mandatory to understand distribution pattern of organisms to address further improvements.8 Preterm infants are susceptible to infection at a higher rate than other patients because of the immature skin barrier, immature and not fully developed immune regulatory systems, prolonged hospitalization, and frequent need for invasive procedures.29 In addition, lack of preexisting T-cell
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
678
Late-Onset Bacterial Sepsis in Preterm Neonates
Shah et al.
679
a
Outcomes
No infection (N ¼ 6,405)
GP infection (N ¼ 909)
GN infection (N ¼ 195)
p-Value None vs. GP
p-Value None vs. GN
p-Value GP vs. GN
Composite outcomea, n (%)
1,979 (31)
552 (61)
153 (78)
< 0.0001
< 0.0001
< 0.0001
Mortality, n (%)
474 (7)
92 (10)
50 (26)
0.0037
< 0.0001
< 0.0001
Bronchopulmonary dysplasia, n (%)
1,059 (19)
357 (43)
85 (56)
< 0.0001
< 0.0001
0.0038
Severe retinopathy of prematurity, n (%)
179 (7)
108 (16)
31 (22)
< 0.0001
< 0.0001
0.4440
Necrotizing enterocolitis, n (%)
218 (4)
125 (14)
42 (22)
< 0.0001
< 0.0001
0.0076
Duration of mechanical ventilation, d, median (interquartile range)
1 (0, 4)
12 (3, 33)
15 (4, 36)
< 0.0001
< 0.0001
0.5481
Duration of respiratory support, d, median (interquartile range)
8 (2, 35)
51 (23, 87)
59 (17, 97)
< 0.0001
< 0.0001
0.8008
Duration of stay, d, median (interquartile range)
34 (14, 58)
73 (46, 107)
79 (39, 108)
< 0.0001
< 0.0001
0.4785
Composite outcome of mortality or bronchopulmonary dysplasia.
Table 3 Adjusted odds ratio for neonatal outcomes Outcomes
Comparison
Adjusted odds ratio (95% CI)a
p-Value
Bronchopulmonary dysplasia or Mortality
Gram-positive vs. no infection
1.44 (1.21, 1.71)
< 0.0001
Gram-negative vs. no infection
2.79 (1.96, 3.97)
< 0.0001
Bronchopulmonary dysplasia
Mortality
Gram-negative vs. gram-positive
1.94 (1.33, 2.83)
0.0005
Gram-positive vs. no infection
1.41 (1.18, 1.68)
0.0001
Gram-negative vs. no infection
1.94 (1.34, 2.82)
0.0005
Gram-negative vs. gram-positive
1.38 (0.93, 2.04)
0.1104
Gram-positive vs. no infection
1.20 (0.90, 1.59)
0.2265
Gram-negative vs. no infection
3.40 (2.24, 5.16)
< 0.0001
Gram-negative vs. gram-positive
2.85 (1.80, 4.49)
< 0.0001
Abbreviations: CI, confidence interval; SNAPII, Score for Acute Neonatal Physiology version II. a Adjusted for gestational age, sex, SNAPII score 20, small for gestational age, central catheter d > 4, total parenteral nutrition d > 7.
memory and transition from a sterile intrauterine environment to antigen rich extrauterine life contributes to already existing higher risk.14 Several studies have evaluated risk factors for LOS. These predictors include low gestational age, low birth weight,5 high severity of illness on admission as identified by higher SNAP score,2 low Apgar score,2 male sex,2 peripheral venous catheters, central catheters,16,17 longer duration of parenteral nutrition,16,17 drainage tubes, maternal chorioamnionitis, and maternal exposure to antibiotics.1,3,12,14 Moreover, recently the role of the gut microbiome and its alteration has been identified as a possible mechanism for bacterial translocation and increased risk for LOS and NEC.30 A higher prevalence of Enterobacteriaceae and Staphylococcaceae in the gut was identified in infants who developed LOS.28 Our study confirmed many of these risk factors as significant predictors for LOS. Several strategies targeting these risk factors such as standardization of central catheter management practices,9,31,32 strict
guidelines regarding hand hygiene,12,14 antibiotic stewardship to rationalize and target the proper use of antimicrobials,12 promoting the use of breastfeeding or breast milk, and in some studies, use of external agents such as lactoferrin30 have shown benefit in the reduction of LOS. We identified higher odds of mortality or BPD among infants with LOS. In addition, NEC and severe ROP were also increased along with length of stay. There was a graded response observed among infants with no infections, grampositive infections, and gram-negative infections. Several other studies have similarly identified higher odds of mortality among infants with LOS due to gram-negative organisms.3,10,18,26,33 The increase in BPD among infants who had LOS has been also reported in previous studies. Exposure to bacteria and inflammatory markers during the antenatal and postnatal periods has been shown to affect lung development by arresting the growth of alveoli and promoting growth of American Journal of Perinatology
Vol. 32
No. 7/2015
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
Table 2 Neonatal outcomes of study infants
Late-Onset Bacterial Sepsis in Preterm Neonates
Shah et al.
Table 4 Organisms and frequency Gram-positive organisms
Gram-negative organisms
Organism
Blood
CSF
Both
Organism
Blood
CSF
Both
Coagulase-negative Staphylococci
703
34
5
Escherichia coli
81
6
1
Coagulase-positive Staphylococcus aureus
84
3
2
Enterobacteriaceae
31
0
2
Enterococci
34
0
4
Klebsiella
31
1
1
Group B streptococcus
27
2
2
Serratia marcescens
14
1
3
Streptococci
7
1
0
Pseudomonas
14
0
0
Pneumococci
1
1
0
Citrobacter
5
0
0
Acinetobacter
1
0
1
Proteus species
2
0
0
Total
179
8
8
Total
856
41
11
Abbreviation: CSF, cerebrospinal fluid.
disruptive fibrosis, which contributes to the development of BPD.34 The proinflammatory cascade which occurs with LOS has been suspected to be associated with inflammation as a precursor for the development of BPD.33 This study also confirms previous reports10,35 of increased length of stay and prolonged ventilation in infants with LOS. As such, it has been suggested that LOS leads to an increase in hospital costs36 and is associated with abnormal neurodevelopmental outcomes.33,37 A large collaborative study from France reported that at a 5-year neurodevelopmental assessment, infants that had LOS in the NICU had higher odds of significant neurodevelopmental disability (OR 1.71, 95% CI 1.14–2.56) compared with those that did not.38 Other studies have similarly reported a higher risk of cerebral palsy (OR 3.23, 95% CI 1.23–8.48)39 and motor impairment, especially in cases of gram-negative sepsis.40 Unfortunately, we did not have long-term neurodevelopmental data from our cohort to evaluate. The strengths of our study include the nearly populationbased sample and collection of contemporary and detailed data for individual risk factors from the antenatal, intrapartum, and postnatal periods. However, we must acknowledge the limitations of our study, which include the fact that sepsis was diagnosed using one blood culture. However, all infants were clinically symptomatic and received antibiotics for longer than 2 days. We did not assess variation between centers in this study as this was not our objective; however, we are aware that variations in prophylactic and therapeutic interventions between centers can influence the rates of LOS.41 During the study period, many of our centers participated in a nationwide collaborative program and implemented interventions to reduce adverse neonatal outcomes with remarkable success.13 In conclusion, LOS, though declining in incidence, remains a major factor associated with mortality and BPD. Among infants with LOS, gram-negative sepsis was associated with the highest risk of mortality or BPD compared with gram-positive sepsis or not having sepsis. Similar and ongoing surveillance is needed to assess variations, changes in epidemiology and type of organAmerican Journal of Perinatology
Vol. 32
No. 7/2015
isms causing LOS. Moreover, such surveillance will identify areas for future improvements and help to develop efforts to reduce adverse consequences for fragile neonates.
Clinical Perspective • LOS is a major risk factor for morbidity and mortality in very preterm neonates. • Infants with gram-negative bacterial sepsis had the highest risk of adverse outcomes as compared with those with gram-positive sepsis or no sepsis. • Correlation of LOS with long-term outcomes is needed.
Site Investigators for the Canadian Neonatal Network Prakesh S Shah (Director, Canadian Neonatal Network), Mount Sinai Hospital, Toronto, Ontario; Adele Harrison, Victoria General Hospital, Victoria, British Columbia; Anne Synnes, British Columbia Children’s Hospital, Vancouver, British Columbia; Zenon Cieslak, Royal Columbian Hospital, New Westminster, British Columbia; Todd Sorokan, Surrey Memorial Hospital, Surrey, British Columbia; Wendy Yee, Foothills Medical Centre, Calgary, Alberta; Khalid Aziz, Royal Alexandra Hospital, Edmonton, Alberta; Zarin Kalapesi, Regina General Hospital, Regina, Saskatchewan; Koravangattu Sankaran, Royal University Hospital, Saskatoon, Saskatchewan; Mary Seshia, Winnipeg Health Sciences Centre, Winnipeg, Manitoba; Ruben Alvaro, St. Boniface General Hospital, Winnipeg, Manitoba; Sandesh Shivananda, Hamilton Health Sciences Centre, Hamilton, Ontario; Orlando Da Silva, London Health Sciences Centre, London, Ontario; Chuks Nwaesei, Windsor Regional Hospital, Windsor, Ontario; Kyong-Soon Lee, Hospital for Sick Children, Toronto, Ontario; Prakesh Shah, Mount Sinai Hospital, Toronto, Ontario; Michael Dunn, Sunnybrook Health Sciences Centre, Toronto, Ontario; Nicole Rouvinez-Bouali, Children’s Hospital of Eastern Ontario and Ottawa General Hospital, Ottawa, Ontario; Kimberly Dow, Kingston General Hospital, Kingston, Ontario; Lajos
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
680
Late-Onset Bacterial Sepsis in Preterm Neonates
Disclosure None of the authors have anything to declare or have any conflicts of interest.
9 Shane AL, Stoll BJ. Neonatal sepsis: progress towards improved
outcomes. J Infect 2014;68(Suppl 1):S24–S32 10 Stoll BJ, Hansen N, Fanaroff AA, et al. Late-onset sepsis in very low
11
12
13
14
15
16
17
Acknowledgments We would like to thank the staff of the CNN Coordinating Centre for providing organizational support to the CNN and Ruth Warre for editorial assistance with this article. The CNN Coordinating Centre and Dr. Warre are based at the Maternal-Infant Care Research Centre at Mount Sinai Hospital, Toronto, Ontario, Canada, which is supported by funding from the Canadian Institutes for Health Research and Mount Sinai Hospital, Toronto.
18
19
20
21
References 1 Didier C, Streicher MP, Chognot D, et al. Late-onset neonatal
2
3
4
5
6
7 8
infections: incidences and pathogens in the era of antenatal antibiotics. Eur J Pediatr 2012;171(4):681–687 Escobar GJ, Li DK, Armstrong MA, et al. Neonatal sepsis workups in infants >/¼2000 grams at birth: A population-based study. Pediatrics 2000;106(2 Pt 1):256–263 Hornik CP, Fort P, Clark RH, et al. Early and late onset sepsis in verylow-birth-weight infants from a large group of neonatal intensive care units. Early Hum Dev 2012;88(Suppl 2):S69–S74 Lim WH, Lien R, Huang YC, et al. Prevalence and pathogen distribution of neonatal sepsis among very-low-birth-weight infants. Pediatr Neonatol 2012;53(4):228–234 Ozkan H, Cetinkaya M, Koksal N, Celebi S, Hacımustafaoglu M. Culture-proven neonatal sepsis in preterm infants in a neonatal intensive care unit over a 7 year period: coagulase-negative Staphylococcus as the predominant pathogen. Pediatr Int 2014; 56(1):60–66 Tsai MH, Hsu JF, Chu SM, et al. Incidence, clinical characteristics and risk factors for adverse outcome in neonates with late-onset sepsis. Pediatr Infect Dis J 2014;33(1):e7–e13 Stoll BJ, Hansen N. Infections in VLBW infants: studies from the NICHD Neonatal Research Network. Semin Perinatol 2003;27(4):293–301 Bizzarro MJ, Raskind C, Baltimore RS, Gallagher PG. Seventy-five years of neonatal sepsis at Yale: 1928-2003. Pediatrics 2005; 116(3):595–602
681
22
23
24
25
26
27
28
birth weight neonates: the experience of the NICHD Neonatal Research Network. Pediatrics 2002;110(2 Pt 1):285–291 Levit O, Bhandari V, Li FY, Shabanova V, Gallagher PG, Bizzarro MJ. Clinical and laboratory factors that predict death in very low birth weight infants presenting with late-onset sepsis. Pediatr Infect Dis J 2014;33(2):143–146 Shane AL, Stoll BJ. Recent developments and current issues in the epidemiology, diagnosis, and management of bacterial and fungal neonatal sepsis. Am J Perinatol 2013;30(2):131–141 Shah PS, Yoon W, Kalapesi Z, Bassil K, Dunn M, Lee SK. Seasonal variations in healthcare-associated infection in neonates in Canada. Arch Dis Child Fetal Neonatal Ed 2013;98(1):F65–F69 Higgins RD, Baker CJ, Raju TN. Executive summary of the workshop on infection in the high-risk infant. J Perinatol 2010;30(6): 379–383 Ehrenkranz RA, Das A, Wrage LA, et al; Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network. Early nutrition mediates the influence of severity of illness on extremely LBW infants. Pediatr Res 2011; 69(6):522–529 Perlman SE, Saiman L, Larson EL. Risk factors for late-onset health care-associated bloodstream infections in patients in neonatal intensive care units. Am J Infect Control 2007;35(3):177–182 Samanta S, Farrer K, Breathnach A, Heath PT. Risk factors for late onset gram-negative infections: a case-control study. Arch Dis Child Fetal Neonatal Ed 2011;96(1):F15–F18 Downey LC, Smith PB, Benjamin DK Jr. Risk factors and prevention of late-onset sepsis in premature infants. Early Hum Dev 2010;86 (Suppl 1):7–12 Sarkar S, Bhagat I, DeCristofaro JD, Wiswell TE, Spitzer AR. A study of the role of multiple site blood cultures in the evaluation of neonatal sepsis. J Perinatol 2006;26(1):18–22 Richardson DK, Gray JE, McCormick MC, Workman K, Goldmann DA. Score for Neonatal Acute Physiology: a physiologic severity index for neonatal intensive care. Pediatrics 1993;91(3): 617–623 Shennan AT, Dunn MS, Ohlsson A, Lennox K, Hoskins EM. Abnormal pulmonary outcomes in premature infants: prediction from oxygen requirement in the neonatal period. Pediatrics 1988;82(4): 527–532 International Committee for the Classification of Retinopathy of Prematurity. The International Classification of Retinopathy of Prematurity revisited. Arch Ophthalmol 2005;123(7): 991–999 Bell MJ, Ternberg JL, Feigin RD, et al. Neonatal necrotizing enterocolitis. Therapeutic decisions based upon clinical staging. Ann Surg 1978;187(1):1–7 Canadian Neonatal NetworkCanadian Neonatal Network Abstractor’s Manual. v 2.1.2, 1–94. 2014. Available at: www.canadianneonatalnetwork.org. Accessed 1 June, 2014 Stoll BJ, Hansen NI, Bell EF, et al; Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network. Neonatal outcomes of extremely preterm infants from the NICHD Neonatal Research Network. Pediatrics 2010;126(3):443–456 Gordon A, Isaacs D. Late onset neonatal Gram-negative bacillary infection in Australia and New Zealand: 1992-2002. Pediatr Infect Dis J 2006;25(1):25–29 Vergnano S, Menson E, Kennea N, et al. Neonatal infections in England: the NeonIN surveillance network. Arch Dis Child Fetal Neonatal Ed 2011;96(1):F9–F14 Stewart CJ, Marrs EC, Magorrian S, et al. The preterm gut microbiota: changes associated with necrotizing enterocolitis and infection. Acta Paediatr 2012;101(11):1121–1127
American Journal of Perinatology
Vol. 32
No. 7/2015
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
Kovacs, Jewish General Hospital, Montréal, Québec; Keith Barrington, Hôpital Sainte-Justine, Montréal, Québec; Bruno Piedboeuf, Centre Hospitalier Universitaire de Québec, Sainte Foy, Patricia Riley, Montréal Children’s Hospital, Montréal, Québec; Daniel Faucher, Royal Victoria Hospital, Montréal, Québec; Valerie Bertelle, Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Québec; Rody Canning, Moncton Hospital, Moncton, New Brunswick; Barbara Bulleid, Dr. Everett Chalmers Hospital, Fredericton, New Brunswick; Cecil Ojah, Saint John Regional Hospital, Saint John, New Brunswick; Wayne Andrews, Janeway Children’s Health and Rehabilitation Centre, St. John’s, Newfoundland; Douglas McMillan, IWK Health Centre, Halifax, Nova Scotia; Andrzej Kajetanowicz, Cape Breton Regional Hospital, Sydney, Nova Scotia; Shoo K Lee (Chairman, Canadian Neonatal Network), Mount Sinai Hospital, Toronto, Ontario.
Shah et al.
Late-Onset Bacterial Sepsis in Preterm Neonates
Shah et al.
29 Bizzarro MJ. Health care-associated infections in the neonatal
30
31
32
33 34
35
intensive care unit: barriers to continued success. Semin Perinatol 2012;36(6):437–444 Sherman MP. New concepts of microbial translocation in the neonatal intestine: mechanisms and prevention. Clin Perinatol 2010;37(3):565–579 Bizzarro MJ, Sabo B, Noonan M, Bonfiglio MP, Northrup V, Diefenbach K; Central Venous Catheter Initiative Committee. A quality improvement initiative to reduce central line-associated bloodstream infections in a neonatal intensive care unit. Infect Control Hosp Epidemiol 2010;31(3): 241–248 Kaplan HC, Lannon C, Walsh MC, Donovan EF; Ohio Perinatal Quality Collaborative. Ohio statewide quality-improvement collaborative to reduce late-onset sepsis in preterm infants. Pediatrics 2011;127(3):427–435 Adams-Chapman I. Long-term impact of infection on the preterm neonate. Semin Perinatol 2012;36(6):462–470 Kramer BW, Kramer S, Ikegami M, Jobe AH. Injury, inflammation, and remodeling in fetal sheep lung after intra-amniotic endotoxin. Am J Physiol Lung Cell Mol Physiol 2002;283(2): L452–L459 Pessoa-Silva CL, Miyasaki CH, de Almeida MF, Kopelman BI, Raggio RL, Wey SB. Neonatal late-onset bloodstream infection: attribut-
American Journal of Perinatology
Vol. 32
No. 7/2015
36
37
38
39
40
41
able mortality, excess of length of stay and risk factors. Eur J Epidemiol 2001;17(8):715–720 Johnson TJ, Patel AL, Jegier BJ, Engstrom JL, Meier PP. Cost of morbidities in very low birth weight infants. J Pediatr 2013; 162(2):243–249, e1 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(19):2357–2365 Mitha A, Foix-L’Hélias L, Arnaud C, et al; EPIPAGE Study Group. Neonatal infection and 5-year neurodevelopmental outcome of very preterm infants. Pediatrics 2013;132(2):e372–e380 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(2):e348–e357 van der Ree M, Tanis JC, Van Braeckel KN, Bos AF, Roze E. Functional impairments at school age of preterm born children with lateonset sepsis. Early Hum Dev 2011;87(12):821–826 Helwich E, Wójkowska-Mach J, Borszewska-Kornacka M, et al. [Epidemiology of infections in very low birth weight infants. Polish Neonatology Network research]. Med Wieku Rozwoj 2013;17(3):224–231
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
682
675
Risk Factors and Outcomes of Late-Onset Bacterial Sepsis in Preterm Neonates Born at < 32 Weeks’ Gestation Jyotsna Shah, MD, CCRP, DTMH1 Ann L. Jefferies, MD, MEd, FRCPC1 Eugene W. Yoon, MSc2 Shoo K. Lee, MBBS, PhD, FRCPC1,2 Prakesh S. Shah, MD, MSc, MRCP, FRCPC1,2 and on behalf of the Canadian Neonatal Network
Toronto, Toronto, Canada 2 Maternal-Infant Care Research Centre, Mount Sinai Hospital, Toronto, Canada
Address for correspondence Jyotsna Shah, MD, CCRP, DTMH, Department of Pediatrics, Mount Sinai Hospital, Rm 19-231, 600 University Avenue, Toronto, ON M5G 1X5, Canada (e-mail: [email protected]).
Am J Perinatol 2015;32:675–682.
Abstract
Keywords
► ► ► ► ► ►
infant premature nosocomial infection late-onset sepsis mortality bronchopulmonary dysplasia
Objective This study aims to identify the incidence, risk factors, and outcomes of lateonset sepsis in preterm neonates in Canadian neonatal intensive care units (NICUs). Study Design This retrospective analysis included preterm infants born at < 32 weeks’ gestation and admitted to 29 NICUs in the Canadian Neonatal Network during the years 2010 and 2011. Infants were classified into three groups: no infection, grampositive infection, and gram-negative infection. Late-onset sepsis was defined as positive blood and/or spinal fluid cultures after 3 days of birth. Risk factors and the primary outcome of mortality or bronchopulmonary dysplasia (BPD) were compared between the groups. Results Out of the 7,509 neonates, 6,405 (85%) had no infection, 909 (12%) had grampositive, and 195 (3%) had gram-negative infections. Lower gestation, higher Score for Neonatal Acute Physiology, version II scores, the presence of central catheters for > 4 days, parenteral nutrition for > 7 days, and prolonged duration of nothing by mouth were associated with late-onset sepsis. After controlling for confounders, the odds ratio (OR) of mortality/BPD were higher in infants who had gram-negative (OR 2.79, 95% confidence interval [CI] 1.96–3.97) and gram-positive (OR 1.44, 95% CI 1.21–1.71) sepsis as compared with no infection. Conclusions Bacterial late-onset sepsis in very preterm neonates was associated with mortality and BPD. Neonates with gram-negative sepsis had the highest risk of adverse outcomes as compared with gram-positive sepsis or no sepsis.
Late-onset sepsis (LOS) is a major cause of mortality and morbidity in the neonatal intensive care unit (NICU). The incidence of LOS is reported to vary from 20 to 50%, depending upon the patient population, with higher rates among lower gestation and lower birth weight infants.1–6 Data from the Neonatal Network of the National Institute of Child Health
and Human Development in the United States reported that LOS occurred in almost 25% of very low-birth-weight infants.7 The epidemiology of the pathogens responsible for LOS has changed in the past 75 years.8 In the early 1950s, the organisms most responsible for neonatal sepsis were Staphylococcus sp., in the 1960s it was gram-negative
received July 10, 2014 accepted after revision August 28, 2014 published online December 8, 2014
Copyright © 2015 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI http://dx.doi.org/ 10.1055/s-0034-1393936. ISSN 0735-1631.
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
1 Department of Pediatrics, Mount Sinai Hospital, University of
Late-Onset Bacterial Sepsis in Preterm Neonates
Shah et al.
organisms, and since the 1970s coagulase-negative Staphylococcus sp. have emerged as the most common sepsis-causing organisms. In recent years, it has been reported that among cases of LOS in very low-birth-weight infants, 20 to 30% are due to gram-negative organisms, whereas as much as 50% of early-onset infections are due to gram-positive organisms.9–11 No increase in the overall rates of early-onset sepsis or LOS has been reported in recent years subsequent to an increased rate of antepartum or intrapartum chemoprophylaxis for group B streptococcus; however, it is suspected that this may have led to the emergence of resistant strains of gram-negative bacteria.1,12 We have previously identified a steady decline in the rate of LOS in Canadian NICUs (from 4.5 infections/1,000 days in 2003 to 3.5/1,000 days in 2009, a 20% reduction)13; however, it is important to monitor and conduct regular surveillance of the outcomes of infants who develop LOS compared with those without sepsis.14 The risk factors for LOS in preterm neonates include extreme prematurity, low birth weight, use of invasive procedures, including intubation, the presence of central lines, need for total parental nutrition with lipids, receipt of broadspectrum antibiotics, and delayed initiation of enteral feed.1,12,15–17 In addition, the morbidity and mortality of neonates with gram-negative sepsis is high. The invasive and fulminant course of certain gram-negative infections in preterm infants is often devastating.18 Recently, Levit et al reported that need for intubation, pressors, hypoglycemia, and thrombocytopenia were associated with mortality among LOS neonates.11 According to a report from the National Institute of Child Health and Human Development (NICHD) network NICUs between 1998 and 2000, there were higher rates of intraventricular hemorrhage, patent ductus arteriosus, necrotizing enterocolitis (NEC), and duration of hospitalization, especially those who had gram-negative sepsis.10 Contemporary information regarding the rates, risk factors, and outcomes of neonates with LOS caused by bacteria, especially gram-negative sepsis, is needed to counsel parents and to develop indicators for further quality improvement, as well as preventative strategies.14 Our objective of this study was to identify the associated risk factors and outcomes of preterm infants born at < 32 weeks’ gestation who developed LOS during their stay in NICUs in Canada.
Patients and Methods Population We conducted a retrospective analysis of preterm neonates born at < 32 weeks’ gestation and admitted to 29 level-3 NICUs participating in the Canadian Neonatal Network (CNN) during the years 2010 and 2011 (2-year period). We excluded infants who were moribund (palliative care planned at birth), infants with major congenital anomalies, and infants who had early-onset sepsis (blood or spinal fluid culture-positive within the first 3 days of birth).
Exposure (Infecting Pathogen) Infants were identified as having LOS when either blood and/ or cerebrospinal fluid cultures were positive for a pathogenic American Journal of Perinatology
Vol. 32
No. 7/2015
organism in a symptomatic infant after 3 days of birth. The cohort was divided into three groups: neonates who had no infections during their stay in a participating NICU, neonates with gram-positive infections, and neonates with gram-negative infections. Infants who had only viral, mycoplasma, ureaplasma, or fungal infections were excluded as the objective of this study was to assess the impact of bacterial infections. Infants who were infected with both gram-positive and gram-negative organisms were assigned to the gramnegative group. The presence of an organism in one blood or spinal fluid culture was adequate to diagnose LOS. Due to the small size of infants, the technical challenges associated with drawing samples for two blood cultures and the fact that a single positive culture is considered adequate for the diagnosis of neonatal sepsis,19 a second positive culture was not mandatory.
Risk Factor Evaluation To assess the risk factors for LOS, the following information was obtained from the CNN database: duration of central catheter (dichotomized at < or 4 days), duration of nothing by mouth, severity of illness using the Score for Acute Neonatal Physiology, version II (SNAPII) score,20 duration of parenteral nutrition (dichotomized at < 7 or 7 days), maternal chorioamnionitis (defined as maternal fever during peripartum period, and either foul smelling liquor, uterine tenderness or high maternal white cell count), maternal antibiotics during labor, and small for gestational age, defined as birth weight below the 10th centile for gestational age and sex. These risk factors and outcomes were compared between the three groups. Dichotomization for the central catheter duration and duration of parenteral nutrition was based on median duration of the respective variables in the network data.
Outcomes The primary outcome of this study was a composite of mortality and bronchopulmonary dysplasia (BPD). Mortality was defined as any neonate who died 3 days or more after birth, but before discharge from the NICU. BPD was defined as supplemental oxygen dependency at 36 weeks corrected age or at the time of transfer to another medical facility if that occurred before 36 weeks’ corrected gestational age.21 The secondary outcomes were BPD, severe (> stage 2) retinopathy of prematurity (ROP), stage 2 or 3 NEC, and the following measures of resource utilization: duration of mechanical ventilation, duration of respiratory support, and length of stay. Gestational age was defined as the best estimate based on obstetric history, obstetric examination, and first prenatal ultrasound examination. Severity of illness was assessed using the SNAPII score. ROP was defined according to the international classification.22 NEC was defined according to the Bell criteria.23 Duration of mechanical ventilation was defined as the total number of days during which the infant was on mechanical ventilation (not including noninvasive modes) during any part of the day. Duration of respiratory support was defined as the duration of invasive respiratory
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
676
Late-Onset Bacterial Sepsis in Preterm Neonates
Shah et al.
677
support, continuous positive pressure airway pressure support or other noninvasive respiratory support. Length of stay was defined as the total number of days that an infant stayed in the NICU before death, discharge to home, or transfer to a level-2 hospital.
Data on individual infants were collected as part of the CNN’s ongoing data collection system. Out of the 30 NICUs in Canada, 29 NICUs were participating in the network during the study period. At all affiliated sites, demographic and outcomes data were collected from patient charts by trained research assistants using a computerized data entry program according to the standardized outcome definitions.24 Data were transmitted electronically to the CNN coordinating center where they were stored. Data were collected on each infant until death or discharge from the NICU. Data collection was approved by each institution’s research ethics board or institutional quality improvement committee as appropriate.
Analyses Statistical analyses were performed with SAS software 9.2 (SAS Institute, Inc., Cary, NC). Baseline demographic characteristics, perinatal risk factors, and the incidence of primary and secondary outcomes among infants in the no infection (reference group), gram-positive infection, and gram-negative infection groups were compared using Pearson chisquare test for categorical variables and Student’s t-test or analysis of variance for continuous variables. Multivariate logistic regression analysis was used to examine the effect of clinically significant and other confounders identified from univariate analyses, which included gestational age, sex, SNAPII 20, small for gestational age, central catheter duration > 4 days, and parenteral nutrition > 7 days.
Results Out of the 33,633 neonates admitted to NICUs in Canada during the study time period, 7,509 neonates were included in the study cohort as described in ►Fig. 1. Of these, 6,405 (85%) neonates had no infections, 909 (12%) had gram-positive infections, and 195 (3%) had gram-negative infections. Of the 1,114 neonates with bacterial LOS, 1,034 (94%) were blood-culture positive, 21 (2%) were both blood-culture and spinal-fluid positive, and 49 (4%) were only spinal-fluid positive for organisms. Of the 1,034 neonates with positive blood culture 583 had spinal fluid evaluated. Comparison of baseline characteristics between the three groups revealed that infants who had infections had a lower gestational age, lower birth weight, and higher severity of illness at admission (SNAPII score > 20). In addition, assessment of risk factors revealed that a higher number of infants with infection had longer durations of central catheter use, parenteral nutrition administration, and nothing by mouth, as well as higher rates of maternal chorioamnionitis (►Table 1). Comparison of the primary outcomes of mortality/BPD and the secondary outcomes revealed that the outcome rates
Fig. 1 Patient selection.
were significantly higher in the infection groups with the worst outcomes occurring in infants with gram-negative infections (►Table 2). Resource utilization data also revealed significantly higher durations of respiratory support and length of stay for infants with infection, especially those with gram-negative infections (►Table 2). After adjusting for confounders and baseline differences, there were significantly higher odds of mortality/BPD among infants with both gram-positive and gram-negative infections as compared with no infection (►Table 3). Comparison of patients with gram-positive infections to those with gramnegative infections revealed that the combined outcome of mortality/BPD, and BPD and mortality individually were significantly higher in infants with gram-negative infections (►Table 3). Organisms and their frequency are reported in ►Table 4. The organisms found most frequently in infants in the grampositive group were coagulase-negative Staphylococci, coagulase-positive Staphylococcus aureus, and Enterococci, whereas in the gram-negative group Escherichia coli, Klebsiella, and Enterobacter were found most frequently (►Table 4).
Discussion In this large population-based cohort of very preterm neonates, we identified that LOS still remains a major cause of mortality and morbidity. Among the infants with bacterial LOS, 80% were caused by gram-positive organisms, whereas 20% were caused by gram-negative organisms. Analysis of the outcomes of infants who had bacterial LOS revealed higher odds of mortality, BPD, severe ROP, and NEC, as well as higher resource utilization compared with infants who did not have sepsis. Lower gestational age, higher SNAPII scores, the presence of central catheters for > 4 days (median duration of the cohort), parenteral nutrition administration for > 7 days (median duration of the cohort), and prolonged duration of nothing by mouth were significant risk factors associated with bacterial LOS. The most common gram-positive American Journal of Perinatology
Vol. 32
No. 7/2015
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
Data Collection
Late-Onset Bacterial Sepsis in Preterm Neonates
Shah et al.
Table 1 Maternal and neonatal characteristics of study infants No infection (N ¼ 6,405)
GP infection (N ¼ 909)
GN infection (N ¼ 195)
p-Value None vs. GP
p-Value None vs. GN
p-Value GP vs. GN
Antenatal steroids, n (%)
5,378 (87)
768 (88)
167 (87)
0.9741
0.8761
0.8745
Multiples, n (%)
1,901 (30)
273 (30)
63 (32)
0.9227
0.4089
0.4762
Maternal chorioamnionitis, n (%)
889 (19)
137 (20)
43 (28)
0.2719
0.0044
0.0465
Vaginal births, n (%)
2,630 (41)
353 (39)
104 (53)
0.1485
0.0005
0.0001
Gestational age (wk), mean (SD)
28.5 (2.2)
26.8 (2.2)
26.4 (2.2)
< 0.0001
< 0.0001
0.0167
Birth weight (g), mean (SD)
1,240 (394)
978 (314)
956 (337)
< 0.0001
< 0.0001
0.4446
Female, n (%)
2,925 (46)
392 (43)
85 (44)
0.1490
0.3704
0.8623
Small for gestational age, n (%)
593 (9)
116 (13)
19 (10)
0.0006
0.9358
0.1401
Apgar score < 7 at 5 min, n (%)
1,716 (27)
349 (39)
71 (37)
< 0.0001
0.0016
0.7201
SNAP score, median (interquartile range)
9 (0, 14)
14 (7, 21)
14 (9, 24)
< 0.0001
< 0.0001
0.1309
SNAPII score 20, n (%)
1,013 (16)
251 (28)
68 (35)
< 0.0001
< 0.0001
0.0559
Duration of central catheter, median (interquartile range)
7 (0, 14)
22 (10, 38)
24 (12, 48)
< 0.0001
< 0.0001
0.2520
Duration of central catheter > 4 d, n (%)
3,735 (58)
789 (87)
180 (92)
< 0.0001
< 0.0001
0.0517
Duration of nothing by mouth during stay, median (interquartile range)
1 (1, 3)
6 (2, 12)
8 (3, 16)
< 0.0001
< 0.0001
0.1161
Duration of parenteral nutrition, median (interquartile range)
10 (6, 16)
28 (19, 45)
30 (15, 53)
< 0.0001
< 0.0001
0.8798
Duration of parenteral nutrition > 7 d, n (%)
4,062 (63)
876 (96)
176 (90)
< 0.0001
< 0.0001
0.0003
Characteristics Maternal/peripartum variables
Birth variables
Neonatal course
Abbreviations: GN, gram-negative; GP, gram-positive; SNAPII, Score for Acute Neonatal Physiology version II.
organism was coagulase-negative Staphylococci and the most common gram-negative organism was E. coli. Several single center and multicenter reports, from developed countries of rates of LOS and the contribution of grampositive and gram-negative organisms, are available with varying estimates of the incidence of LOS that range from 25 to 35% in very low-birth-weight infants.4,5,25 In these reports, gram-positive organisms accounted for 55 to 65% of the cases, gram-negative organisms for 25 to 35% of the cases, and fungi, mycoplasma, ureaplasma, and, rarely, viral organisms for the remaining cases.3,4,26 Rates of gram-negative organisms have been reported to be as high as 46%.27 Among the gram-positive organisms, coagulase-negative Staphylococci has been reported to be responsible for > 50% of LOS in preterm infants5; however, the distribution of gramnegative organisms is variable. Pseudomonas,5,6 Klebsiella,4,27 E. coli, and Enterobacter species17,28 have been reported to be the predominant organisms from various units. In our cohort, American Journal of Perinatology
Vol. 32
No. 7/2015
among the infants with bacterial LOS, gram-negative organisms accounted for 20% of the cases, slightly lower than previous reports from other networks and single centers. We speculate that this reduction is a result of our networkwide quality improvement efforts over 2009 to 2012 and suggests a continued decline in rates of LOS.13 However, there was no statistically significant difference in the rate of infection between two years of this study period. Our rate of coagulase-negative Staphylococci was higher than reported in other studies, which could be because we have not mandated two cultures in the diagnosis of sepsis. Ongoing surveillance like this report is mandatory to understand distribution pattern of organisms to address further improvements.8 Preterm infants are susceptible to infection at a higher rate than other patients because of the immature skin barrier, immature and not fully developed immune regulatory systems, prolonged hospitalization, and frequent need for invasive procedures.29 In addition, lack of preexisting T-cell
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
678
Late-Onset Bacterial Sepsis in Preterm Neonates
Shah et al.
679
a
Outcomes
No infection (N ¼ 6,405)
GP infection (N ¼ 909)
GN infection (N ¼ 195)
p-Value None vs. GP
p-Value None vs. GN
p-Value GP vs. GN
Composite outcomea, n (%)
1,979 (31)
552 (61)
153 (78)
< 0.0001
< 0.0001
< 0.0001
Mortality, n (%)
474 (7)
92 (10)
50 (26)
0.0037
< 0.0001
< 0.0001
Bronchopulmonary dysplasia, n (%)
1,059 (19)
357 (43)
85 (56)
< 0.0001
< 0.0001
0.0038
Severe retinopathy of prematurity, n (%)
179 (7)
108 (16)
31 (22)
< 0.0001
< 0.0001
0.4440
Necrotizing enterocolitis, n (%)
218 (4)
125 (14)
42 (22)
< 0.0001
< 0.0001
0.0076
Duration of mechanical ventilation, d, median (interquartile range)
1 (0, 4)
12 (3, 33)
15 (4, 36)
< 0.0001
< 0.0001
0.5481
Duration of respiratory support, d, median (interquartile range)
8 (2, 35)
51 (23, 87)
59 (17, 97)
< 0.0001
< 0.0001
0.8008
Duration of stay, d, median (interquartile range)
34 (14, 58)
73 (46, 107)
79 (39, 108)
< 0.0001
< 0.0001
0.4785
Composite outcome of mortality or bronchopulmonary dysplasia.
Table 3 Adjusted odds ratio for neonatal outcomes Outcomes
Comparison
Adjusted odds ratio (95% CI)a
p-Value
Bronchopulmonary dysplasia or Mortality
Gram-positive vs. no infection
1.44 (1.21, 1.71)
< 0.0001
Gram-negative vs. no infection
2.79 (1.96, 3.97)
< 0.0001
Bronchopulmonary dysplasia
Mortality
Gram-negative vs. gram-positive
1.94 (1.33, 2.83)
0.0005
Gram-positive vs. no infection
1.41 (1.18, 1.68)
0.0001
Gram-negative vs. no infection
1.94 (1.34, 2.82)
0.0005
Gram-negative vs. gram-positive
1.38 (0.93, 2.04)
0.1104
Gram-positive vs. no infection
1.20 (0.90, 1.59)
0.2265
Gram-negative vs. no infection
3.40 (2.24, 5.16)
< 0.0001
Gram-negative vs. gram-positive
2.85 (1.80, 4.49)
< 0.0001
Abbreviations: CI, confidence interval; SNAPII, Score for Acute Neonatal Physiology version II. a Adjusted for gestational age, sex, SNAPII score 20, small for gestational age, central catheter d > 4, total parenteral nutrition d > 7.
memory and transition from a sterile intrauterine environment to antigen rich extrauterine life contributes to already existing higher risk.14 Several studies have evaluated risk factors for LOS. These predictors include low gestational age, low birth weight,5 high severity of illness on admission as identified by higher SNAP score,2 low Apgar score,2 male sex,2 peripheral venous catheters, central catheters,16,17 longer duration of parenteral nutrition,16,17 drainage tubes, maternal chorioamnionitis, and maternal exposure to antibiotics.1,3,12,14 Moreover, recently the role of the gut microbiome and its alteration has been identified as a possible mechanism for bacterial translocation and increased risk for LOS and NEC.30 A higher prevalence of Enterobacteriaceae and Staphylococcaceae in the gut was identified in infants who developed LOS.28 Our study confirmed many of these risk factors as significant predictors for LOS. Several strategies targeting these risk factors such as standardization of central catheter management practices,9,31,32 strict
guidelines regarding hand hygiene,12,14 antibiotic stewardship to rationalize and target the proper use of antimicrobials,12 promoting the use of breastfeeding or breast milk, and in some studies, use of external agents such as lactoferrin30 have shown benefit in the reduction of LOS. We identified higher odds of mortality or BPD among infants with LOS. In addition, NEC and severe ROP were also increased along with length of stay. There was a graded response observed among infants with no infections, grampositive infections, and gram-negative infections. Several other studies have similarly identified higher odds of mortality among infants with LOS due to gram-negative organisms.3,10,18,26,33 The increase in BPD among infants who had LOS has been also reported in previous studies. Exposure to bacteria and inflammatory markers during the antenatal and postnatal periods has been shown to affect lung development by arresting the growth of alveoli and promoting growth of American Journal of Perinatology
Vol. 32
No. 7/2015
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
Table 2 Neonatal outcomes of study infants
Late-Onset Bacterial Sepsis in Preterm Neonates
Shah et al.
Table 4 Organisms and frequency Gram-positive organisms
Gram-negative organisms
Organism
Blood
CSF
Both
Organism
Blood
CSF
Both
Coagulase-negative Staphylococci
703
34
5
Escherichia coli
81
6
1
Coagulase-positive Staphylococcus aureus
84
3
2
Enterobacteriaceae
31
0
2
Enterococci
34
0
4
Klebsiella
31
1
1
Group B streptococcus
27
2
2
Serratia marcescens
14
1
3
Streptococci
7
1
0
Pseudomonas
14
0
0
Pneumococci
1
1
0
Citrobacter
5
0
0
Acinetobacter
1
0
1
Proteus species
2
0
0
Total
179
8
8
Total
856
41
11
Abbreviation: CSF, cerebrospinal fluid.
disruptive fibrosis, which contributes to the development of BPD.34 The proinflammatory cascade which occurs with LOS has been suspected to be associated with inflammation as a precursor for the development of BPD.33 This study also confirms previous reports10,35 of increased length of stay and prolonged ventilation in infants with LOS. As such, it has been suggested that LOS leads to an increase in hospital costs36 and is associated with abnormal neurodevelopmental outcomes.33,37 A large collaborative study from France reported that at a 5-year neurodevelopmental assessment, infants that had LOS in the NICU had higher odds of significant neurodevelopmental disability (OR 1.71, 95% CI 1.14–2.56) compared with those that did not.38 Other studies have similarly reported a higher risk of cerebral palsy (OR 3.23, 95% CI 1.23–8.48)39 and motor impairment, especially in cases of gram-negative sepsis.40 Unfortunately, we did not have long-term neurodevelopmental data from our cohort to evaluate. The strengths of our study include the nearly populationbased sample and collection of contemporary and detailed data for individual risk factors from the antenatal, intrapartum, and postnatal periods. However, we must acknowledge the limitations of our study, which include the fact that sepsis was diagnosed using one blood culture. However, all infants were clinically symptomatic and received antibiotics for longer than 2 days. We did not assess variation between centers in this study as this was not our objective; however, we are aware that variations in prophylactic and therapeutic interventions between centers can influence the rates of LOS.41 During the study period, many of our centers participated in a nationwide collaborative program and implemented interventions to reduce adverse neonatal outcomes with remarkable success.13 In conclusion, LOS, though declining in incidence, remains a major factor associated with mortality and BPD. Among infants with LOS, gram-negative sepsis was associated with the highest risk of mortality or BPD compared with gram-positive sepsis or not having sepsis. Similar and ongoing surveillance is needed to assess variations, changes in epidemiology and type of organAmerican Journal of Perinatology
Vol. 32
No. 7/2015
isms causing LOS. Moreover, such surveillance will identify areas for future improvements and help to develop efforts to reduce adverse consequences for fragile neonates.
Clinical Perspective • LOS is a major risk factor for morbidity and mortality in very preterm neonates. • Infants with gram-negative bacterial sepsis had the highest risk of adverse outcomes as compared with those with gram-positive sepsis or no sepsis. • Correlation of LOS with long-term outcomes is needed.
Site Investigators for the Canadian Neonatal Network Prakesh S Shah (Director, Canadian Neonatal Network), Mount Sinai Hospital, Toronto, Ontario; Adele Harrison, Victoria General Hospital, Victoria, British Columbia; Anne Synnes, British Columbia Children’s Hospital, Vancouver, British Columbia; Zenon Cieslak, Royal Columbian Hospital, New Westminster, British Columbia; Todd Sorokan, Surrey Memorial Hospital, Surrey, British Columbia; Wendy Yee, Foothills Medical Centre, Calgary, Alberta; Khalid Aziz, Royal Alexandra Hospital, Edmonton, Alberta; Zarin Kalapesi, Regina General Hospital, Regina, Saskatchewan; Koravangattu Sankaran, Royal University Hospital, Saskatoon, Saskatchewan; Mary Seshia, Winnipeg Health Sciences Centre, Winnipeg, Manitoba; Ruben Alvaro, St. Boniface General Hospital, Winnipeg, Manitoba; Sandesh Shivananda, Hamilton Health Sciences Centre, Hamilton, Ontario; Orlando Da Silva, London Health Sciences Centre, London, Ontario; Chuks Nwaesei, Windsor Regional Hospital, Windsor, Ontario; Kyong-Soon Lee, Hospital for Sick Children, Toronto, Ontario; Prakesh Shah, Mount Sinai Hospital, Toronto, Ontario; Michael Dunn, Sunnybrook Health Sciences Centre, Toronto, Ontario; Nicole Rouvinez-Bouali, Children’s Hospital of Eastern Ontario and Ottawa General Hospital, Ottawa, Ontario; Kimberly Dow, Kingston General Hospital, Kingston, Ontario; Lajos
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
680
Late-Onset Bacterial Sepsis in Preterm Neonates
Disclosure None of the authors have anything to declare or have any conflicts of interest.
9 Shane AL, Stoll BJ. Neonatal sepsis: progress towards improved
outcomes. J Infect 2014;68(Suppl 1):S24–S32 10 Stoll BJ, Hansen N, Fanaroff AA, et al. Late-onset sepsis in very low
11
12
13
14
15
16
17
Acknowledgments We would like to thank the staff of the CNN Coordinating Centre for providing organizational support to the CNN and Ruth Warre for editorial assistance with this article. The CNN Coordinating Centre and Dr. Warre are based at the Maternal-Infant Care Research Centre at Mount Sinai Hospital, Toronto, Ontario, Canada, which is supported by funding from the Canadian Institutes for Health Research and Mount Sinai Hospital, Toronto.
18
19
20
21
References 1 Didier C, Streicher MP, Chognot D, et al. Late-onset neonatal
2
3
4
5
6
7 8
infections: incidences and pathogens in the era of antenatal antibiotics. Eur J Pediatr 2012;171(4):681–687 Escobar GJ, Li DK, Armstrong MA, et al. Neonatal sepsis workups in infants >/¼2000 grams at birth: A population-based study. Pediatrics 2000;106(2 Pt 1):256–263 Hornik CP, Fort P, Clark RH, et al. Early and late onset sepsis in verylow-birth-weight infants from a large group of neonatal intensive care units. Early Hum Dev 2012;88(Suppl 2):S69–S74 Lim WH, Lien R, Huang YC, et al. Prevalence and pathogen distribution of neonatal sepsis among very-low-birth-weight infants. Pediatr Neonatol 2012;53(4):228–234 Ozkan H, Cetinkaya M, Koksal N, Celebi S, Hacımustafaoglu M. Culture-proven neonatal sepsis in preterm infants in a neonatal intensive care unit over a 7 year period: coagulase-negative Staphylococcus as the predominant pathogen. Pediatr Int 2014; 56(1):60–66 Tsai MH, Hsu JF, Chu SM, et al. Incidence, clinical characteristics and risk factors for adverse outcome in neonates with late-onset sepsis. Pediatr Infect Dis J 2014;33(1):e7–e13 Stoll BJ, Hansen N. Infections in VLBW infants: studies from the NICHD Neonatal Research Network. Semin Perinatol 2003;27(4):293–301 Bizzarro MJ, Raskind C, Baltimore RS, Gallagher PG. Seventy-five years of neonatal sepsis at Yale: 1928-2003. Pediatrics 2005; 116(3):595–602
681
22
23
24
25
26
27
28
birth weight neonates: the experience of the NICHD Neonatal Research Network. Pediatrics 2002;110(2 Pt 1):285–291 Levit O, Bhandari V, Li FY, Shabanova V, Gallagher PG, Bizzarro MJ. Clinical and laboratory factors that predict death in very low birth weight infants presenting with late-onset sepsis. Pediatr Infect Dis J 2014;33(2):143–146 Shane AL, Stoll BJ. Recent developments and current issues in the epidemiology, diagnosis, and management of bacterial and fungal neonatal sepsis. Am J Perinatol 2013;30(2):131–141 Shah PS, Yoon W, Kalapesi Z, Bassil K, Dunn M, Lee SK. Seasonal variations in healthcare-associated infection in neonates in Canada. Arch Dis Child Fetal Neonatal Ed 2013;98(1):F65–F69 Higgins RD, Baker CJ, Raju TN. Executive summary of the workshop on infection in the high-risk infant. J Perinatol 2010;30(6): 379–383 Ehrenkranz RA, Das A, Wrage LA, et al; Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network. Early nutrition mediates the influence of severity of illness on extremely LBW infants. Pediatr Res 2011; 69(6):522–529 Perlman SE, Saiman L, Larson EL. Risk factors for late-onset health care-associated bloodstream infections in patients in neonatal intensive care units. Am J Infect Control 2007;35(3):177–182 Samanta S, Farrer K, Breathnach A, Heath PT. Risk factors for late onset gram-negative infections: a case-control study. Arch Dis Child Fetal Neonatal Ed 2011;96(1):F15–F18 Downey LC, Smith PB, Benjamin DK Jr. Risk factors and prevention of late-onset sepsis in premature infants. Early Hum Dev 2010;86 (Suppl 1):7–12 Sarkar S, Bhagat I, DeCristofaro JD, Wiswell TE, Spitzer AR. A study of the role of multiple site blood cultures in the evaluation of neonatal sepsis. J Perinatol 2006;26(1):18–22 Richardson DK, Gray JE, McCormick MC, Workman K, Goldmann DA. Score for Neonatal Acute Physiology: a physiologic severity index for neonatal intensive care. Pediatrics 1993;91(3): 617–623 Shennan AT, Dunn MS, Ohlsson A, Lennox K, Hoskins EM. Abnormal pulmonary outcomes in premature infants: prediction from oxygen requirement in the neonatal period. Pediatrics 1988;82(4): 527–532 International Committee for the Classification of Retinopathy of Prematurity. The International Classification of Retinopathy of Prematurity revisited. Arch Ophthalmol 2005;123(7): 991–999 Bell MJ, Ternberg JL, Feigin RD, et al. Neonatal necrotizing enterocolitis. Therapeutic decisions based upon clinical staging. Ann Surg 1978;187(1):1–7 Canadian Neonatal NetworkCanadian Neonatal Network Abstractor’s Manual. v 2.1.2, 1–94. 2014. Available at: www.canadianneonatalnetwork.org. Accessed 1 June, 2014 Stoll BJ, Hansen NI, Bell EF, et al; Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network. Neonatal outcomes of extremely preterm infants from the NICHD Neonatal Research Network. Pediatrics 2010;126(3):443–456 Gordon A, Isaacs D. Late onset neonatal Gram-negative bacillary infection in Australia and New Zealand: 1992-2002. Pediatr Infect Dis J 2006;25(1):25–29 Vergnano S, Menson E, Kennea N, et al. Neonatal infections in England: the NeonIN surveillance network. Arch Dis Child Fetal Neonatal Ed 2011;96(1):F9–F14 Stewart CJ, Marrs EC, Magorrian S, et al. The preterm gut microbiota: changes associated with necrotizing enterocolitis and infection. Acta Paediatr 2012;101(11):1121–1127
American Journal of Perinatology
Vol. 32
No. 7/2015
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
Kovacs, Jewish General Hospital, Montréal, Québec; Keith Barrington, Hôpital Sainte-Justine, Montréal, Québec; Bruno Piedboeuf, Centre Hospitalier Universitaire de Québec, Sainte Foy, Patricia Riley, Montréal Children’s Hospital, Montréal, Québec; Daniel Faucher, Royal Victoria Hospital, Montréal, Québec; Valerie Bertelle, Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Québec; Rody Canning, Moncton Hospital, Moncton, New Brunswick; Barbara Bulleid, Dr. Everett Chalmers Hospital, Fredericton, New Brunswick; Cecil Ojah, Saint John Regional Hospital, Saint John, New Brunswick; Wayne Andrews, Janeway Children’s Health and Rehabilitation Centre, St. John’s, Newfoundland; Douglas McMillan, IWK Health Centre, Halifax, Nova Scotia; Andrzej Kajetanowicz, Cape Breton Regional Hospital, Sydney, Nova Scotia; Shoo K Lee (Chairman, Canadian Neonatal Network), Mount Sinai Hospital, Toronto, Ontario.
Shah et al.
Late-Onset Bacterial Sepsis in Preterm Neonates
Shah et al.
29 Bizzarro MJ. Health care-associated infections in the neonatal
30
31
32
33 34
35
intensive care unit: barriers to continued success. Semin Perinatol 2012;36(6):437–444 Sherman MP. New concepts of microbial translocation in the neonatal intestine: mechanisms and prevention. Clin Perinatol 2010;37(3):565–579 Bizzarro MJ, Sabo B, Noonan M, Bonfiglio MP, Northrup V, Diefenbach K; Central Venous Catheter Initiative Committee. A quality improvement initiative to reduce central line-associated bloodstream infections in a neonatal intensive care unit. Infect Control Hosp Epidemiol 2010;31(3): 241–248 Kaplan HC, Lannon C, Walsh MC, Donovan EF; Ohio Perinatal Quality Collaborative. Ohio statewide quality-improvement collaborative to reduce late-onset sepsis in preterm infants. Pediatrics 2011;127(3):427–435 Adams-Chapman I. Long-term impact of infection on the preterm neonate. Semin Perinatol 2012;36(6):462–470 Kramer BW, Kramer S, Ikegami M, Jobe AH. Injury, inflammation, and remodeling in fetal sheep lung after intra-amniotic endotoxin. Am J Physiol Lung Cell Mol Physiol 2002;283(2): L452–L459 Pessoa-Silva CL, Miyasaki CH, de Almeida MF, Kopelman BI, Raggio RL, Wey SB. Neonatal late-onset bloodstream infection: attribut-
American Journal of Perinatology
Vol. 32
No. 7/2015
36
37
38
39
40
41
able mortality, excess of length of stay and risk factors. Eur J Epidemiol 2001;17(8):715–720 Johnson TJ, Patel AL, Jegier BJ, Engstrom JL, Meier PP. Cost of morbidities in very low birth weight infants. J Pediatr 2013; 162(2):243–249, e1 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(19):2357–2365 Mitha A, Foix-L’Hélias L, Arnaud C, et al; EPIPAGE Study Group. Neonatal infection and 5-year neurodevelopmental outcome of very preterm infants. Pediatrics 2013;132(2):e372–e380 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(2):e348–e357 van der Ree M, Tanis JC, Van Braeckel KN, Bos AF, Roze E. Functional impairments at school age of preterm born children with lateonset sepsis. Early Hum Dev 2011;87(12):821–826 Helwich E, Wójkowska-Mach J, Borszewska-Kornacka M, et al. [Epidemiology of infections in very low birth weight infants. Polish Neonatology Network research]. Med Wieku Rozwoj 2013;17(3):224–231
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
682
