+ MODEL

Pediatrics and Neonatology (2016) xx, 1e6

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: http://www.pediatr-neonatol.com

ORIGINAL ARTICLE

Ventilator-Associated Pneumonia in Low

Birth Weight Neonates at a Neonatal
Intensive Care Unit: A Retrospective
Observational Study
Pei-Lun Lee a, Wei-Te Lee a, Hsiu-Lin Chen a,b,*

a
Department of Pediatrics, Kaohsiung Medical University Hospital, Kaohsiung, Taiwan
b
Department of Respiratory Therapy, College of Medicine, Kaohsiung Medical University, Kaohsiung,
Taiwan

Received Jul 6, 2015; received in revised form Oct 7, 2015; accepted Oct 30, 2015
Available online - - -

Key Words Background: Ventilator-associated pneumonia (VAP) is one of the most common healthcare-
intubation; associated infections among ventilated patients. The aim of this study was to determine the
low birth weight clinical characteristics and risk factors for the development of VAP in intubated low birth
infant; weight (LBW) neonates in a neonatal intensive care unit.
neonatal intensive Methods: LBW infants (<2.5 kg) admitted to the neonatal intensive care unit of Kaohsiung
care unit; Medical University Hospital from January 2005 to December 2009 were enrolled. We retro-
preterm; spectively analyzed perinatal and neonatal data of the enrolled intubated LBW infants by
ventilator associated chart review.
pneumonia Results: Six hundred and five LBW infants were analyzed. One hundred and fourteen of the
infants were intubated for >48 hours, 15 (13.2%) of whom had VAP. Of these 15 patients, the
average age at onset of VAP was 24.0
11.2 days, the average postmenstrual age was
30.6
1.8 weeks, and the mean gestational age was 27.1
2.3 weeks, which was signifi-
cantly lower than the mean gestational age in the group without VAP (30.2
3.5 weeks).
The mean birth body weight was 944.4
268.4 g in the VAP group and 1340.1
455.4 g in
the group without VAP (p < 0.001). Longer duration of intubation (odds ratio: 1.35, 95% con-
fidence interval: 1.12e1.62) and parenteral nutrition (odds ratio: 1.32, 95% confidence inter-
val: 1.14e1.51) were found in the VAP group after adjusting for gestational age and birth
weight.
Conclusion: VAP was a problem for the LBW infants with intubation for >48 hours in
our neonatal intensive care unit. VAP most frequently occurred at a postmenstrual age of
30e32 weeks in this study. Longer duration of tube placement and parenteral nutrition were

* Corresponding author. Department of Pediatrics, Kaohsiung Medical University Hospital, Number 100, Tzyou 1st Road, Kaohsiung City
807, Taiwan.
E-mail address: 840062@ms.kmuh.org.tw (H.-L. Chen).

http://dx.doi.org/10.1016/j.pedneo.2015.10.014
1875-9572/Copyright ª 2016, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-
NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article in press as: Lee P-L, et al., Ventilator-Associated Pneumonia in Low Birth Weight Neonates at a Neonatal Intensive
Care Unit: A Retrospective Observational Study, Pediatrics and Neonatology (2016), http://dx.doi.org/10.1016/j.pedneo.2015.10.014
 + MODEL
2 P.-L. Lee et al

found in the VAP group. Early removal of the endotracheal tube and adequate enteral nutri-
tion may decrease the occurrence of VAP in LBW infants.
Copyright ª 2016, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/
by-nc-nd/4.0/).

1. Introduction
Ventilator-associated pneumonia (VAP) is a common and
severe complication of critical illnesses in intensive care
units (ICUs). It is associated with longer durations of hos-
pital and ICU stay and high rates of morbidity and mortality
in patients in an ICU. Several studies have reported on VAP
in pediatric patients; however, few studies have focused on
neonates, especially those with a birth weight <2.5 kg (low
birth weight, LBW).1e5 The leading causes of death in ne-
onates are disorders related to short gestation and LBW.6
Respiratory distress syndrome is the most common respi-
ratory diagnosis in preterm LBW neonates requiring respi-
ratory support.7 In addition, the incidence of respiratory
distress syndrome has been reported to be higher among
infants with a lower gestational age and lower birth
weight.8 To rescue these smaller preterm infants, respira-
tory support including intubation and the use of mechanical
ventilation is crucial after birth. Neonatal pneumonia is
also a common cause of respiratory distress in neonates,
and it also necessitates mechanical ventilation after birth.9
Although the survival of preterm LBW neonates has
improved with improvements in the techniques of neonatal
mechanical ventilation and advancing neonatal care, a
longer duration of mechanical ventilation can still lead to
respiratory morbidity including VAP and bronchopulmonary
dysplasia.10,11 Therefore, VAP in preterm infants, and
especially LBW infants, is an important issue.
To evaluate the condition of VAP in our neonatal ICU
(NICU), we enrolled intubated LBW infants undergoing
mechanical ventilation for >48 hours from January 2005 to
December 2009. The aim of this study was to determine the
clinical characteristics and risk factors for the development
of VAP in LBW neonates in our NICU.
discharge from the hospital. The NTISS score was recorded
within 48 hours of admission. The neonates intubated for
>48 hours were divided into two groups: those with VAP
(VAP group) and those without VAP (no-VAP group). Clinical
symptoms, laboratory findings, radiological signs, and
microbiological findings of the VAP patients were recorded.
The diagnosis of VAP was made on the basis of new or
progressive respiratory symptoms and infiltrates/patches
on chest X ray. According to Centers for Disease Control and
Prevention (CDC) diagnostic criteria for VAP in infants
younger than 1 year, the diagnosis should be based on all of
the three items of radiologic signs, clinical signs and
symptoms, and microbiological findings.13 Tracheal aspirate
in each enrolled patient was obtained for microbiological
data. VAP was documented from the chart review. X rays
were interpreted by the authors.
2.2. Statistical analysis
Categorical data are presented as numbers (%) and
continuous data as the mean
standard deviation. The
characteristics of the patients with and without VAP were
compared using Student t tests for numerical data and c2
tests for categorical data. Univariate and multivariate an-
alyses were performed to explore the risk factors related to
VAP. JMP version 10 (SAS Institute Inc., Cary, NC, USA) was
used for all statistical analyses. The level of statistical
significance was defined as p < 0.05.
2.3. Ethics
This study was approved by the Institutional Review Board
of Kaohsiung Medical University Hospital: KMUHIRBe
20140028.

2. Methods 3. Results

2.1. Patients
From January 2005 to December 2009, all intubated LBW
infants (birth weight <2.5 kg) admitted to the NICU of
Kaohsiung Medical University Hospital were enrolled. We
retrospectively analyzed perinatal and neonatal data of the
intubated LBW infants undergoing mechanical ventilation
for >48 hours. The perinatal data included gestational age,
body weight, percentage of small for gestational age, Apgar
score at 1 and 5 minutes, National Therapeutic Intervention
Scoring System (NTISS) scores, endotracheal intubation,
and length of hospital stay. Small for gestational age was
defined as a weight below the 10th percentile for the
gestational age.12 Death was regarded as death before
From January 2005 to December 2009, 662 LBW neonates
were admitted to the NICU of Kaohsiung Medical University
Hospital. One hundred and fifteen (17.4%) of the LBW neo-
nates required intubation after birth; one of whom was
excluded from this study because of an intubation period
<48 hours. Of the remaining 114 patients who received
48 hours of mechanical ventilation, 15 patients (13.2%) met
the criteria of VAP and had received antibiotics for at least
7 days. The characteristics of the intubated LBW neonates
with and without VAP are detailed in Table 1. The mean
gestational age was 27.1
2.3 weeks in the VAP group, which
was significantly lower than that in the no-VAP group
(30.2
3.5 weeks). The mean birth body weight was
944.4
268.4 g in the VAP group compared with

Please cite this article in press as: Lee P-L, et al., Ventilator-Associated Pneumonia in Low Birth Weight Neonates at a Neonatal Intensive
Care Unit: A Retrospective Observational Study, Pediatrics and Neonatology (2016), http://dx.doi.org/10.1016/j.pedneo.2015.10.014
 + MODEL
Ventilator-associated pneumonia in neonates 3

Table 1 Characteristics of the intubated LBW neonates with and without VAP.
Characteristics VAP (n Z 15) No-VAP (n Z 99) p
Male*, n (%) 10 (66.7) 57 (57.6) 0.505
Gestational age,y weeks 27.1
2.3 30.2
3.5 <0.001
Birth body weight,y g 944.4
268.4 1340.1
455.4 <0.001
Birth weight group* <0.00
1501e2500 g, n (%) 0 (0) 35 (35.4)
1001e1500 g, n (%) 4 (3.7) 40 (40.4)
<1001 g, n (%) 11 (73.3) 24 (24.2)
Apgar score at 1 miny 5.5
1.4 5.1
1.9 0.336
Apgar score at 5 miny 6.8
1.3 6.9
1.8 0.721
NTISS scorey 24.3
6.0 20.6
5.7 0.024
Patent ductus arteriosus,* n (%) 14 (93.3) 71 (71.7) 0.073
Breast feeding,* n (%) 14 (93.3) 87 (87.9) 0.536
Duration of intubation, dy 58.7
39.6 12.6
19.2 <0.001
PICC, n (%)* 15 (100) 90 (90.9) 0.224
TPN, n (%)* 15 (100) 85 (85.9) 0.120
Duration of TPN,y d 85.3
45.9 30.9
25.9 <0.001
Length of stay in hospital,y d 107.3
55.2 44.8
32.9 <0.001
Death,* n (%) 2 (13.3) 17 (17.2) 0.710
Data are presented as n (%) or mean
standard deviation.
LBW Z low birth weight; NTISS Z National Therapeutic Intervention Scoring System; PICC Z peripheral inserted central catheter;
TPN Z total parenteral nutrition; VAP Z ventilator-associated pneumonia.
* c2 test.
y
t test.

1340.1
455.4 g in the no-VAP group (p < 0.001). Of the 15 pathogens from the remaining 12 patients were Klebsiella
patients with VAP, 11 (73.3%) weighed <1001 g. There were pneumoniae (13.3%) and Burkholderia cepacia (13.3%)
no significant differences in Apgar scores at 1 and 5 minutes (Table 3). The most prevalent clinical symptoms and lab-
between the two groups. The VAP group had a significantly oratory findings were new onset of apnea (n Z 10, 66.7%)
higher NTISS score than the no-VAP group (24.3
6.0 vs. and an elevated C-reactive protein level (>5 mg/L) (n Z 9,
20.6
5.7, p Z 0.024). The rates of breast feeding, use of 60%) (Table 4). The infection rate was 7.1 pneumonia in-
peripheral inserted central catheters, and use of total fections per 1000 ventilator days.
parenteral nutrition (TPN) were similar in both groups.
However, the duration of TPN was longer in the VAP group
(85.3
45.9 d vs. 30.9
25.9 d, p < 0.001). The VAP group Table 2 Characteristics of the 15 infants with VAP.
also had a significantly longer duration of intubation
(58.7
39.6 d vs. 12.6
19.2 d, p < 0.001) and length of Patient BBW GA Postnatal PMA at onset Duration of
no. (g) (wk) age at of VAP (PMA ventilation
hospital stay compared with the no-VAP group (107.3
55.2 d
vs. 44.8
32.9 d, p < 0.001). The mortality rate was similar in onset of in wk) (d)
both groups (13.3% vs. 17.2%, p Z 0.710). VAP (d)
Table 2 shows the postnatal and postmenstrual age (PMA) 1 853 25þ3 26 29þ1 94
at the onset of VAP, and the duration of mechanical venti- 2 695 27þ3 12 29þ1 33
lation of the 15 infants with VAP. The mean postnatal age and 3 541 27þ3 21 30þ3 34
PMA at the onset of VAP in these 15 LBW neonates were 4 717 25þ6 26 29þ4 89
24.0
11.2 days and 30.6
1.8 weeks, respectively. Two 5 1420 29þ5 11 31þ2 3
patients (Cases 7 and 9) in the VAP group died before 6 850 25þ4 42 31þ4 44
discharge. Their postnatal ages at onset of VAP were 17 days 7 1280 29 17 31þ3 160
and 14 days, and they died at 160 days and 109 days of age, 8 880 26þ2 33 31 31
respectively. The cause of death in both patients was bron- 9 940 26þ2 14 28þ2 109
chopulmonary dysplasia with pulmonary hypertension. 10 810 27 44 33þ2 41
Table 3 and Table 4 show the clinical characteristics of 11 1000 25þ5 20 28þ4 36
the 15 patients with VAP. All of the VAP patients had pos- 12 880 26 37 31þ2 47
itive findings on chest X-ray with deterioration in respira- 13 1470 33þ2 7 34þ2 49
tory condition. Endotracheal aspirate was used for 14 1100 28 27 31þ6 39
bacterial cultures in all of the patients. Three (20%) 15 730 24þ5 23 28 71
patients were diagnosed based on new or progressive BBW Z birth body weight; GA Z gestational age;
infiltrates and respiratory symptoms without confirmed PMA Z postmenstrual age; VAP Z ventilator-associated
cultured organisms. The most commonly cultured pneumonia.

Please cite this article in press as: Lee P-L, et al., Ventilator-Associated Pneumonia in Low Birth Weight Neonates at a Neonatal Intensive
Care Unit: A Retrospective Observational Study, Pediatrics and Neonatology (2016), http://dx.doi.org/10.1016/j.pedneo.2015.10.014
 + MODEL
4 P.-L. Lee et al

score was associated with the occurrence of VAP [odds ratio

Table 3 Microorganisms recovered from sputum cultures
(OR): 1.11, 95% confidence interval (CI): 1.01e1.20], as
of 15 LBW infants with VAP.
were longer durations of intubation and TPN (OR: 1.43, 95%
Organisms n (%) CI: 1.22e1.73 and OR: 1.33, 95% CI: 1.20e1.60, respec-
Klebsiella pneumoniae 2 (13.3) tively). We then performed multivariate analysis to identify
Burkholderia cepacia 2 (13.3) the factors associated with VAP in the intubated LBW ne-
Chryseobacterium meningosepticum 1 (6.7) onates. Longer duration of intubation (OR: 1.35, 95% CI:
Enterobacter cloacae 1 (6.7) 1.12e1.62) and TPN (OR: 1.32, 95% CI: 1.14e1.51) were
Enterococcus 1 (6.7) found in the VAP group after adjusting for gestational age
Escherichia coli 1 (6.7) and birth weight (Table 6).
Pseudomonas aeruginosa 1 (6.7)
Serratia marcescens 1 (6.7)
Stenotrophomonas maltophilia 1 (6.7) 4. Discussion
Staphylococcus haemolyticus 1 (6.7)
No growth 3 (20) In this study, the incidence of VAP was 13.2% in intubated
LBW Z low birth weight; VAP Z ventilator-associated LBW infants on the basis of new or progressive infiltrates
pneumonia. and respiratory symptoms, and the VAP rate was 7.1 epi-
sodes per 1000 ventilator days. VAP is a problem for LBW
infants with endotracheal tubes in a NICU, and its diagnosis
remains a challenge. The most significant risk factor related
Table 4 Clinical symptoms and laboratory findings of 15 to VAP was the duration of intubation. Although the post-
LBW infants with VAP. natal age at the onset of VAP was variable, we found that
Clinical symptoms and laboratory findings n (%) the PMA at onset of VAP was similar.
VAP is often considered to occur after respiratory
New-onset apnea 10 (66.7)
deterioration during ventilatory support, but few studies
Elevated CRP (>5 mg/L) 9 (60.0)
have demonstrated the timing of the onset of VAP among
Bradycardia (<100 beats/min) 5 (33.3)
premature infants. Tekin et al14 reported that the median
Increased respiratory secretion 5 (33.3)
time to a diagnosis of VAP was 32.11
29.3 days after
New-onset tachypnea 3 (20.0)
admission. Our results showed that VAP occurred mainly at
I/T ratio 0.2 2 (13.3)
a PMA of 30e32 weeks. Alveoli are present at 32 weeks of
Thrombocytopenia (<100,000 U/L) 4 (26.7)
age, and the lungs are capable of respiration because of the
Fever (>38 C) 1 (6.7)
alveolocapillary membrane.15 Entering the alveolar period
CRP Z C-reactive protein; I/T Z immature to total neutrophil may be one of the causes of VAP in smaller preterm infants
ratio; LBW Z low birth weight; VAP Z ventilator-associated receiving mechanical ventilatory support.
pneumonia.
Healthcare-associated infections are associated with in-
hospital mortality and morbidity. In addition to bloodstream
and urinary tract infections, VAP is one of the most common
Table 5 Univariate analysis of factors associated with VAP healthcare-associated infections among ventilated patients.
in the intubated LBW neonates. The reported rate of VAP in neonates ranges from 2.7 epi-
sodes to 10.9 episodes per 1000 ventilator days.14,16e20 A
Factors OR 95% CI p
recent report from the International Healthcare-associated
Male, n (%) 1.47 0.48e5.02 0.507 Infection Control Consortium collected data on healthcare-
Gestational age, wk 0.69 0.53e0.86 0.003 associated infections from ICUs worldwide from 2007 to
Birth body weight, g 0.997 0.995e0.999 0.0034 2012.21 The reported mean VAP rate in LBW infants at Level III
Apgar score at 1 min 1.13 0.84e1.58 0.424 ICUs was 8.95 episodes per 1000 ventilator days, which is
Apgar score at 5 min 0.95 0.71e1.33 0.770
NTISS score 1.11 1.01e1.20 0.028
Patent ductus arteriosus 5.52 1.03e102.4 0.107
Table 6 Multivariate logistic regression analysis of factors
Breast feeding 1.93 0.34e36.5 0.542
associated with VAP in intubated LBW neonates.
Duration of intubation, wk 1.43 1.22e1.73 <0.001
Duration of TPN, wk 1.33 1.20e1.60 <0.001 Factors Adjusted OR* 95% CI p
CI Z confidence interval; LBW Z low birth weight; NTISS score 1.07 0.95e1.20 0.298
NTISS Z National Therapeutic Intervention Scoring System; Duration of intubation 1.35 1.12e1.62 0.002
OR Z odds ratio; TPN Z total parenteral nutrition; (additional wk)
VAP Z ventilator-associated pneumonia. Duration of TPN 1.32 1.14e1.51 <0.001
(additional wk)
We further analyzed the factors associated with VAP *Adjusted OR; gestational age and birth weight were adjusted.
using univariate analysis (Table 5). Gender, Apgar score, CI Z confidence interval; LBW Z low birth weight;
NTISS Z National Therapeutic Intervention Scoring System;
and breast feeding were not correlated with VAP, whereas
OR Z odds ratio; TPN Z total parenteral nutrition;
increasing gestational age and birth weight had significant
VAP Z ventilator-associated pneumonia.
effects on lowering the occurrence of VAP. A higher NTISS

Please cite this article in press as: Lee P-L, et al., Ventilator-Associated Pneumonia in Low Birth Weight Neonates at a Neonatal Intensive
Care Unit: A Retrospective Observational Study, Pediatrics and Neonatology (2016), http://dx.doi.org/10.1016/j.pedneo.2015.10.014
 + MODEL
Ventilator-associated pneumonia in neonates
similar to our results. Kawanishi et al22 compared the VAP
rate between 7-day and 14-day ventilator circuit changes,
and found no significant difference (9.66 vs. 8.08 episodes
per 1000 ventilator days). We change the ventilator circuit
every 7 days in our NICU as routine practice, and the rate of
VAP (7.1 episodes per 1000 ventilator days) was slightly lower
than in the study of Kawanishi et al.22
VAP is thought to be associated with lower gestational age
and lower birth weight due to the higher probability of intu-
bation after birth. Geffers et al18 reported 8677 very low birth
weight infants from 52 neonatology departments in Germany.
They found that 41.4% of the neonates who developed
healthcare-associated infections weighed <1 kg, and that a
lower gestational age was also a risk factor for neonatal
mortality. The NTISS score was designed to evaluate the
severity of neonatal illnesses.23 Our recently published study
demonstrated that the NTISS score at 48 hours can effectively
predict mortality in very low birth weight infants.24 Hsu et al25
reported bloodstream infections in 793 neonates, and they
found that NTISS scores 23 had a higher risk (OR: 6.61, 95%
CI: 2.40e26.47) of treatment failure in neonatal bacteremia.
In the current study, the NTISS score in the LBW infants with
VAP was significantly higher than in those without VAP. This
suggests that a higher severity of neonatal illness (requiring
intubation) in LBW neonates after admission increases the
possibility of VAP occurring later during hospitalization.
Patent ductus arteriosus (PDA) is a congenital cardiac
condition in which the neonatal ductus arteriosus fails to
close after birth, and preterm infants tend to have a higher
incidence of PDA. Despite advances in the treatment of
PDA, hemodynamic and pulmonary alterations often occur
in patients with PDA, and the need for respiratory support
may be greater in neonates with PDA. Gonzalez et al26 re-
ported a relationship between PDA and infection, and they
found that infection adversely influenced PDA. Although
there was higher percentage of PDA in the VAP group
compared with the no-VAP group in the current study (Table
1), it did not reach statistical significance (p Z 0.073).
The most common microorganisms contributing to VAP are
Gram-negative bacilli,1,16,27 as also seen in our results. Other
studies have shown that, in addition to Gram-negative bac-
teria, coagulase-negative staphylococci are commonly
involved in VAP, accounting for 13.5e23% of all cases of
VAP.1,16,18,20 Although microbiological findings are essential
to make a diagnosis of VAP according to the CDC diagnostic
criteria for VAP, some authors have found it difficult to
distinguish respiratory infections from colonizers via endo-
tracheal aspirate.28,29 In Germany, Geffers et al17 studied
8677 very low birth weight (VLBW) infants in whom VAP was
diagnosed based on respiratory compromise, radiological
findings and clinical presentation without requiring micro-
biological reports. In our study, three patients were diag-
nosed with VAP based on a deteriorating respiratory
condition, clinical symptoms and new findings on chest X-ray.
One review article concluded that the diagnosis of VAP in
pediatric patients was difficult because of the low sensitivity
of qualitative tracheal aspirate cultures, which are most
commonly used in pediatric ICUs.30 In order to be more
applicable to a clinical setting when diagnosing VAP in pedi-
atric patients, Mariki et al31 proposed modified microbio-
logical criteria combining an endotracheal specimen growth
of one or more organisms and moderate/many
Please cite this article in press as: Lee P-L, et al., Ventilator-Associated Pneumonia in Low Birth Weight Neonates at a Neonatal Intensive
Care Unit: A Retrospective Observational Study, Pediatrics and Neonatology (2016), http://dx.doi.org/10.1016/j.pedneo.2015.10.014
5
polymorphonuclear leukocytes. They found that only 28% of
the 102 patients meeting the inclusion criteria also met both
clinical and radiological CDC criteria for VAP.
Neonatal patients with VAP commonly present with an
unstable temperature, bradycardia, tachycardia, and res-
piratory symptoms.13 In addition to signs of dyspnea, apnea
is more common among neonates with VAP compared with
older children. Apisarnthanarak et al16 reported that hy-
pothermia (79%) and new-onset tachypnea (63%) were the
most common clinical symptoms in extremely preterm ne-
onates. Khattab et al32 compared neonates with and
without VAP, and they found that chest auscultation find-
ings and mucopurulent endotracheal tube secretion
occurred in w80% of the neonates with VAP. In our patients
with VAP, new-onset apnea was the most common clinical
sign. Although apnea is not uncommon, it should be
considered a clue for the diagnosis of VAP in a NICU.
One systemic review showed that pediatric patients with
risk factors including genetic syndrome, steroid therapy,
reintubation, bloodstream infections, prior antibiotic ther-
apy, and bronchoscopy tended to have a higher incidence of
VAP.2 In the neonatal period, the duration of intubation and
previous bloodstream infections were more commonly
associated with VAP. The duration of mechanical ventilation
is reported to be one of the most common risk factors leading
to the occurrence of VAP.2,33e35 In our study, we found longer
duration of intubation and TPN in the VAP group after
multivariate analysis. If the endotracheal tube can be
removed earlier, the risk for VAP is lower. Also, poor nutri-
tional status is associated with difficulty of weaning from
ventilator. We think that durations of intubation and TPN,
rather than just outcome, are important for intubated LBW
infants. Physicians should evaluate the need for intubation
with mechanical ventilation carefully and attempt to
improve the nutritional condition for intubated LBW infants.
Therefore, extubation and introducing enteral feeding as
soon as possible may reduce the risk of VAP. One adult study
demonstrated that enteral feeding may increase the risk of
stomach colonization and lead to a higher incidence of
nosocomial pneumonia.36 However, the early introduction of
parenteral or enteral nutrition has been reported to be well-
tolerated in very preterm infants.37 Early restart or not dis-
continuing enteral feeding may be a strategy to reduce the
occurrence of VAP in intubated LBW neonates.
In conclusion, VAP is a problem for LBW infants with
intubation for >48 hours in our NICU. Longer duration of
intubation and TPN were found in the VAP group. The early
removal of the endotracheal tube and adequate enteral
nutrition may decrease the occurrence of VAP in intubated
LBW neonates.
Conflict of interest
All authors declare that there are no conflicts of interest.
Acknowledgments
The authors are grateful to the staff and members of the
NICU at Kaohsiung Medical University Hospital, Kaohsiung
City, Taiwan.
 + MODEL
6 P.-L. Lee et al

References 20. Cernada M, Aguar M, Brugada M, Gutiérrez A, López JL,

Castell M, et al. Ventilator-associated pneumonia in newborn
infants diagnosed with an invasive bronchoalveolar lavage
1. Srinivasan R, Asselin J, Gildengorin G, Wiener-Kronish J,
technique: a prospective observational study. Pediatr Crit
Flori HR. A prospective study of ventilator-associated pneu-
Care Med 2013;14:55e61.
monia in children. Pediatrics 2009;123:1108e15.
21. Rosenthal VD, Maki DG, Mehta Y, Leblebicioglu H, Memish ZA,
2. Liu B, Li SQ, Zhang SM, Xu P, Zhang X, Zhang YH, et al. Risk
Al-Mousa HH, et al. International nosocomial infection control
factors of ventilator-associated pneumonia in pediatric inten-
consortium (INICC) report, data summary of 43 countries for
sive care unit: a systematic review and meta-analysis. J Thorac
2007e2012. Device-associated module. Am J Infect Control
Dis 2013;5:525e31.
2014;42:942e56.
3. Gautam A, Ganu SS, Tegg OJ, Andresen DN, Wilkins BH,
22. Kawanishi F, Yoshinaga M, Morita M, Shibata Y, Yamada T,
Schell DN. Ventilator-associated pneumonia in a tertiary pae-
Ooi Y, et al. Risk factors for ventilator-associated pneumonia in
diatric intensive care unit: a 1-year prospective observational
neonatal intensive care unit patients. J Infect Chemother
study. Crit Care Resusc 2012;14:283e9.
2014;20:627e30.
4. Foglia E, Meier MD, Elward A. Ventilator-associated pneumonia
23. Gray JE, Richardson DK, McCormick MC, Workman-Daniels K,
in neonatal and pediatric intensive care unit patients. Clin
Goldmann DA. Neonatal therapeutic intervention scoring sys-
Microbiol Rev 2007;20:409e25.
tem: a therapy-based severity-of-illness index. Pediatrics
5. Elward AM, Warren DK, Fraser VJ. Ventilator-associated
1992;90:561e7.
pneumonia in pediatric intensive care unit patients: risk fac-
24. Wu PL, Lee WT, Lee PL, Chen HL. Predictive power of serial
tors and outcomes. Pediatrics 2002;109:758e64.
neonatal therapeutic intervention scoring system scores for
6. Heron M. Deaths: leading causes for 2010. Natl Vital Stat Rep
short-term mortality in very-low-birth-weight infants. Pediatr
2013;62:1e96.
Neonatol 2015;56:108e13.
7. Ramanathan R, Sardesai S. Lung protective ventilatory strate-
25. Hsu JF, Chu SM, Huang YC, Lien R, Huang HR, Lee CW, et al.
gies in very low birth weight infants. J Perinatol 2008;28:
Predictors of clinical and microbiological treatment failure in
S41e6.
neonatal bloodstream infections. Clin Microbiol Infect 2015;
8. Ghafoor T, Mahmud S, Ali S, Dogar SA. Incidence of respiratory
21:482.e9ee17.
distress syndrome. J Coll Physicians Surg Pak 2003;13:271e3.
26. Gonzalez A, Sosenko IR, Chandar J, Hummler H, Claure N,
9. Edwards MO, Kotecha SJ, Kotecha S. Respiratory distress of the
Bancalari E. Influence of infection on patent ductus arteriosus
term newborn infant. Paediatr Respir Rev 2013;14:29e36.
and chronic lung disease in premature infants weighing 1000
10. Johnson JD, Malachowski NC, Grobstein R, Welsh D, Daily WJ,
grams or less. J Pediatr 1996;128:470e8.
Sunshine P. Prognosis of children surviving with the aid of
27. Hocevar SN, Edwards JR, Horan TC, Morrell GC, Iwamoto M,
mechanical ventilation in the newborn period. J Pediatr 1974;
Lessa FC. Device-associated infections among neonatal inten-
84:272e6.
sive care unit patients: incidence and associated pathogens
11. Battin MR, Knight DB, Kuschel CA, Howie RN. Improvement in
reported to the national healthcare safety network,
mortality of very low birthweight infants and the changing
2006e2008. Infect Control Hosp Epidemiol 2012;33:1200e6.
pattern of neonatal mortality: the 50-year experience of one
28. Baltimore RS. The difficulty of diagnosing ventilator-associated
perinatal centre. J Paediatr Child Health 2012;48:596e9.
pneumonia. Pediatrics 2003;112:1420e1.
12. Hsieh WS, Wu HC, Jeng SF, Liao HF, Su YN, Lin SJ, et al.
29. Cordero L, Ayers LW, Miller RR, Seguin JH, Coley BD. Surveil-
Nationwide singleton birth weight percentiles by gestational
lance of ventilator-associated pneumonia in very-low-birth-
age in Taiwan, 1998e2002. Acta Paediatr Taiwan 2006;47:
weight infants. Am J Infect Control 2002;30:32e9.
25e33.
30. Venkatachalam V, Hendley JO, Willson DF. The diagnostic
13. Cernada M, Brugada M, Golombek S, Vento M. Ventilator-
dilemma of ventilator-associated pneumonia in critically ill
associated pneumonia in neonatal patients: an update.
children. Pediatr Crit Care Med 2011;12:286e96.
Neonatology 2014;105:98e107.
31. Mariki P, Rellosa N, Wratney A, Stockwell D, Berger J, Song X,
14. Tekin R, Dal T, Pirinccioglu H, Oygucu SE. A 4-year surveillance
et al. Application of a modified microbiologic criterion for
of device-associated nosocomial infections in a neonatal
identifying pediatric ventilator-associated pneumonia. Am J
intensive care unit. Pediatr Neonatol 2013;54:303e8.
Infect Control 2014;42:1079e83.
15. Moore KL, Persaud TVN. The developing human: clinical ori-
32. Soliman W, Khattab A, El-Lahony D. Ventilator-associated
ented embryology. 7th ed. Philadelphia: Saunders; 2003.
pneumonia in the neonatal intensive care unit. Menoufia Med J
p. 248e9.
2014;27:73.
16. Apisarnthanarak A, Holzmann-Pazgal G, Hamvas A, Olsen MA,
33. Oygur N, Ongun H, Saka O. Risk prediction using a neonatal
Fraser VJ. Ventilator-associated pneumonia in extremely pre-
therapeutic intervention scoring system in VLBW and ELBW
term neonates in a neonatal intensive care unit: characteris-
preterm infants. Pediatr Int 2012;54:496e500.
tics, risk factors, and outcomes. Pediatrics 2003;112:1283e9.
34. Tan B, Zhang F, Zhang X, Huang YL, Gao YS, Liu X, et al. Risk
17. Leistner R, Piening B, Gastmeier P, Geffers C, Schwab F.
factors for ventilator-associated pneumonia in the neonatal
Nosocomial infections in very low birthweight infants in
intensive care unit: a meta-analysis of observational studies.
Germany: current data from the national surveillance sys-
Eur J Pediatr 2014;173:427e34.
tem NEO-KISS. Klin Padiatr 2013;225:75e80.
35. Yuan TM, Chen LH, Yu HM. Risk factors and outcomes for
18. Geffers C, Baerwolff S, Schwab F, Gastmeier P. Incidence of
ventilator-associated pneumonia in neonatal intensive care
healthcare-associated infections in high-risk neonates: results
unit patients. J Perinat Med 2007;35:334e8.
from the German surveillance system for very-low-birthweight
36. Memish ZA, Cunningham G, Oni GA, Djazmati W. The incidence
infants. J Hosp Infect 2008;68:214e21.
and risk factors of ventilator-associated pneumonia in a Riyadh
19. Urzedo JE, Levenhagen MM, Pedroso RS, Abdallah VO,
hospital. Infect Control Hosp Epidemiol 2000;21:271e3.
Sabino SS, Brito DV. Nosocomial infections in a neonatal
37. Su BH. Optimizing nutrition in preterm infants. Pediatr Neo-
intensive care unit during 16 years: 1997e2012. Rev Soc Bras
natol 2014;55:5e13.
Med Trop 2014;47:321e6.

Please cite this article in press as: Lee P-L, et al., Ventilator-Associated Pneumonia in Low Birth Weight Neonates at a Neonatal Intensive
Care Unit: A Retrospective Observational Study, Pediatrics and Neonatology (2016), http://dx.doi.org/10.1016/j.pedneo.2015.10.014