G Model

JIPH-795; No. of Pages 5 ARTICLE IN PRESS

Journal of Infection and Public Health xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Journal of Infection and Public Health

journal homepage: http://www.elsevier.com/locate/jiph

Healthcare associated infections in neonatal intensive care unit and

its correlation with environmental surveillance
Sanjay Kumar a,∗ , Binoy Shankar b , Sugandha Arya a , Manorma Deb c , Harish Chellani a
a
Department of Pediatrics, VMMC & Safdarjung Hospital, New Delhi, India
b
Department of Paediatrics, PGIMER & Dr. RML Hospital, New Delhi, India
c
Department of Microbiology, VMMC & Safdarjung Hospital, New Delhi, India

a r t i c l e i n f o a b s t r a c t

Article history: Healthcare-associated infections (HAI) are frequent complications in neonatal intensive care units (NICU)
Received 5 February 2017 with varying risk factors and bacteriological profile. There is paucity of literature comparing the bacte-
Received in revised form 11 July 2017 riological profile of organisms causing HAI with the environmental surveillance isolates. Therefore, this
Accepted 7 August 2017
study aimed to evaluate demographic profile, risk factors and outcome of HAI in NICU and correlate with
environmental surveillance.
Keywords:
Three hundred newborns with signs and symptoms of sepsis were enrolled in the study group and
Healthcare associated infection
their profile, risk factors and outcome were compared with the control group. Univariate analysis and
Neonatal intensive care unit
Environmental surveillance
multivariable logistic regression were performed. Environmental surveillance results were compared to
the bacteriological profile of HAIs.
We identified lower gestational age, male gender and apgar score less than 7 at 5 min, use of peripheral
vascular catheter & ventilator along with their duration as significant risk factors. Mortality rate was 29%
in the study group (p < 0.05). The HAI site distribution showed blood-stream infections (73%) to be the
most common followed by pneumonia (12%) and meningitis (10%). Gram positive cocci were the most
common isolates in HAI as well as environmental surveillance.
The bacteriological profile of HAI correlates with the environmental surveillance report thus insisting
for periodic surveillance and thereby avoiding irrational antibiotic usage.
© 2017 The Authors. Published by Elsevier Limited on behalf of King Saud Bin Abdulaziz University
for Health Sciences. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction
Healthcare associated infections (HAIs) are of increasing con-
cern in neonatal intensive care units (NICUs) due to advances
in invasive therapeutic and diagnostic procedures and increased
survival of preterm babies. It is defined as localized or systemic
condition resulting from adverse reaction to the presence of an
infectious agent(s) or its toxin(s) and that was not incubating at
the time of admission to the hospital [1]. Most authors describe
HAIs as equivalent to late-onset, or infection after the first 72 h of
life [2]. Reported rates of HAIs per admission in the literature range
from 6% to 50% with 3-to 20-fold higher rates in developing coun-
tries as compared to developed countries [3]. The overall mortality
∗ Corresponding author at: E13, IInd Floor, Defence Colony, New Delhi 110024,
India.
E-mail addresses: drsanjaykumar.am@gmail.com (S. Kumar),
binoypmch@gmail.com (B. Shankar), sugandha arya@hotmail.com (S. Arya),
manormad@gmail.com (M. Deb), chellaniharish@gmail.com (H. Chellani).
rate varies between 20% and 80% depending on the risk factors [4].
In addition, there exists wide variation in the bacteriological pro-
file and antibiogram of microorganisms in different NICUs which
changes consistently with time.
Surveillance of HAI is an essential part of quality and safe patient
care. There are few reports of National Healthcare Safety Network
(NHSN) surveillance in neonatal intensive care units and none in
developing countries [5]. However, hospital based infection control
policy is present in different hospitals in India with the constituting
members responsible for surveillance of infections and providing
methods of control. Surveillance specimens include clinical mate-
rial and environmental samples.
The NICU environmental surfaces harbor large number of
bacterial and fungal taxa associated with nosocomial infection
in neonates. These genera contain many species commensal in
healthy humans but do not necessarily represent pathogenic
strains. Bokulich et al. [6] isolated several taxa, Staphylococcus and
Streptococcus from the NICU surfaces and simultaneously identi-
fied members of these two genera too responsible for the most

http://dx.doi.org/10.1016/j.jiph.2017.08.005
1876-0341/© 2017 The Authors. Published by Elsevier Limited on behalf of King Saud Bin Abdulaziz University for Health Sciences. 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: Kumar S, et al. Healthcare associated infections in neonatal intensive care unit and its correlation
with environmental surveillance. J Infect Public Health (2017), http://dx.doi.org/10.1016/j.jiph.2017.08.005
G Model
JIPH-795; No. of Pages 5 ARTICLE IN PRESS
2 S. Kumar et al. / Journal of Infection and Public Health xxx (2017) xxx–xxx
prevalent cause of the infectious outbreaks in the NICU. But failed
to demonstrate species specific causal association between envi-
ronmental contamination and neonatal infection.
This study therefore, aims to evaluate demographic characteris-
tics, risk factors and bacteriological profile of healthcare associated
infection in NICU and correlate with the environmental surveil-
lance.
Methods
An observational study was conducted over one-year period
with neonates born and investigated for late-onset sepsis during
the hospital stay included in the study group. A total number of
three hundred babies were enrolled on odd days in the study group
based on the prevalence of late-onset sepsis at 20% [7], confidence
interval of 95% and margin of error at 5%. Neonates presenting with
signs of infection with first 72 h of life and those who were dis-
charged home & later presented with sepsis were excluded from
the study. Another three hundred babies who neither developed
sign and symptoms of sepsis nor investigated for late-onset sep-
sis during their hospital course were enrolled consecutively for the
control group.
The study was done at the NICU of a tertiary care teaching
hospital in India with approximately sixteen to twenty thousand
deliveries occurring per year. It is an inborn unit with admission
limited to babies born in the same hospital. The NICU comprises
of two sub units: low risk (level II) and high risk (level III). The
total number of beds is 50 (25 in each) with average bed occu-
pancy rate of 40–50 babies per day. Our hospital has Infection
Control team (HICT). The team comprises of heads from all the
clinical departments, head of microbiology, representatives from
the hospital administration and two infection control (IC) nurses.
The IC policy for NICU includes following: standard precautions,
proper hand rub/hand wash guidelines, minimal invasive pro-
cedure, proper disinfection and sterilization, biomedical waste
disposal and surveillance.
A data collection sheet was prepared separately for the study
and control groups mentioning birth weight, gestation age, gender,
mode of delivery, small for gestation age, resuscitation require-
ment, Apgar score at 1 and 5 min less than 7 and presence of
co-morbidities such as hyaline membrane disease and congenital
malformation. The follow up data was collected on the invasive
device usage (peripheral vascular catheter and ventilation), dura-
tion of device usage, and duration of hospital stay.
The study group patients were followed and categorized into
site specific health-care associated infections using the standard
definitions proposed by the Centre for Disease Control and Preven-
tion [1]. The required samples (blood, cerebrospinal fluid, urine,
endotracheal tube aspirates etc.) were collected using aseptic pre-
cautions and processed using standard methods [8,9]. Blood sample
were analyzed for sepsis screen {Absolute neutrophil count, Imma-
ture to total neutrophil ratio (I/T ratio), ␮ESR, C- reactive protein,
Peripheral smear for sepsis} and blood culture. 1–2 ml of neonate’s
®
blood was taken in BacT/Alert culture media and incubated in
the instrument. From day 2 onwards, when the machine beeps for
growth, bottle was taken out after scanning as positive. It was fol-
lowed by subculture on blood agar, chocolate agar and MacConkey
agar. Plates were incubated at 37 ◦ C aerobically and examined for
bacterial growth. Finally, gram stain, catalase and coagulase test
were done for identification and Kirby Bauer disc diffusion tech-
nique for antibiotic sensitivity. Cerebrospinal spinal fluid (CSF)
samples were analyzed for cell count and protein & glucose esti-
mation. Gram’s stain and latex agglutination was done to identify
bacteria. 1–5 drops of CSF were inoculated on culture media and
incubated for 24 h. Any bacterial growth noted was further tested
Please cite this article in press as: Kumar S, et al. Healthcare associated infections in neonatal intensive care unit and its correlation
with environmental surveillance. J Infect Public Health (2017), http://dx.doi.org/10.1016/j.jiph.2017.08.005
for identification and antibiotic sensitivity. Urine samples were col-
lected by either suprapubic aspiration or catheterization technique
and underwent routine & microscopic examination and culture
& sensitivity analysis. Skin scrapings, endotracheal tube, ear dis-
charges and conjunctival secretions were collected only in the
applicable study subjects for culture and sensitivity. As per pre-
vailing antibiotic policy, Cefotaxime and Amikacin was used as the
first line drug as per hospital policy and switched to second line
drugs Vancomycin or ciprofloxacin if there was no improvement
in 48 h. Meropenem, piperacillin-tazobactum and Linezolide were
used as third line drugs after culture and sensitivity report or in
consultation with microbiologist.
Regular surveillance was carried out in the NICU every four to
six weeks. IC nurses take surveillance specimens from both the
clinical and environmental surfaces. Environmental surfaces reg-
ularly screened in the NICU included: phototherapy lights, radiant
warmers, baby cot, weighing machine, suction tubing’s, laryngo-
scopes, feeding trays, thermometers, feeding counter, air etc. The
surfaces of inanimate articles were sampled with sterile cotton-
tipped swabs. The swabs were moistened with sterile phosphate
buffered saline and streaked across a 4-in2 area (or the entire
surface for surfaces of 4 in2 ); swabs were rotated to ensure full
contact of all parts of the swab tip and the surface. Simultane-
ously culture plates were exposed to the room environment of
both the NICU for one hour. The collected samples were sent to the
microbiology laboratory for analysis. In case of significant growth
from any environmental sample, proper advice was given to the
NICU staff for disinfection. Subsequent sampling was done from the
contaminated article after proper disinfection measures to ensure
decontamination. If the increase in the isolation percentage during
a specified period matched with the monthly growth of that organ-
ism in the surveillance, it was presumed to have some association.
Crude excess mortality was calculated as the difference between
the crude overall case fatality rate of patients with an HAI and the
crude case fatality rate of patients without an HAI in the NICU dur-
ing the same time. Extra length of stay (LOS) was defined as the
difference in median LOS between patients with an HAI and those
without an HAI.
The Institutional ethical committee of Vardhmann Mahavir
Medical College and Safdarjung Hospital ethically approved the
study. Informed consent was waived for this observational study
as there was no deviation from the routine medical practice.
Statistical analysis
Statistical analysis was performed by the SPSS program for
Windows, version 17.0. Quantitative and qualitative variables
are expressed in mean ± SD and relative frequencies respectively.
Odds’ ratio with 95% confidence interval (CI) was calculated for
all variables responsible for causing HAI. p Value was calculated
using parametric and non-parametric tests (t-test, Wilcoxon test
and Pearson’s chi-square test). All risk variables except for inter-
ventional risk factors were subjected to multivariate analysis.
Multivariate logistic regression values were expressed in adjusted
odds’ ration with 95% CI.
Results
Three babies in the study group and two babies in the control left
against medical advice and were not considered in the evaluation.
Thus, the analysis was carried out of 298 babies in the study and
297 babies in the control group.
G Model
JIPH-795; No. of Pages 5 ARTICLE IN PRESS
S. Kumar et al. / Journal of Infection and Public Health xxx (2017) xxx–xxx 3

Table 1 variable [Table 2]. Plotting the receiver operating curve for these
General characteristic of the study and control groups.
three variables showed area under curve 0.67 [95% CI (0.62–0.70)].
Characteristic Study group Control group In the interventional risk category, peripheral vascular catheter
Total number 297 298 (PVC) use (OR 15.39) was shown to have highest association fol-
Sex lowed by mechanical ventilation (MV) use (OR 5.56). The mean
Male (%) 205 (69) 185 (62) difference in the duration of peripheral vascular catheter and ven-
Female (%) 92 (31) 113 (38) tilator use is 2.53 days and 1.3 days respectively (p value <0.0001).
Mode of delivery
The crude excess case fatality rate was 50% and extra length of
Vaginal (%) 218 (73) 203 (68)
Cesarean (%) 79 (27) 95 (32) hospital stay in the study group was calculated as 17 days.
Birth weight
<1500 g (%) 82 (27) 25 (8)
≥1500 g (%) 215 (73) 273 (92) Percentage of NICU HAIs
Gestation age
<37 weeks (%) 161 (54) 121 (41)
≥37 weeks (%) 136 (46) 177 (59)
Out of 297 in the study group, 110 (37%) babies developed clin-
Small for gestation age (%) 72 (24) 80 (27) ical sepsis, 106 (36%) babies had culture positive sepsis, 36 (12%)
Apgar score at 1 min < 7 (%) 57 (19) 62 (20) babies had pneumonia, 30 (10%) meningitis and 6 (2%) had skin
Apgar score at 5 min < 7 (%) 12 (4) 5 (2) & soft tissue lesion, 8 (2.6%) had conjunctivitis and 1 (0.4%) had
Resuscitation required (%) 78 (26) 81 (27)
urinary tract infection.
Hyaline membrane disease (%) 85 (28) 31 (10)
Malformation (%) 9 (3) 11 (4)
Invasive device usage
Peripheral vascular catheter 254 (85) 81 (27) Microbiological profile and Antibiogram
Ventilator 43 (15) 9 (3)
No invasive device usage None 208 (70) The bacteriological profile causing healthcare-associated infec-
Duration of peripheral vascular 4.78 ± 2.44 2.25 ± 1.32
tions in the study group shows predominance of Gram positive
catheter mean ± SD (in days)
Duration of ventilation mean ± SD (in 3.09 ± 0.98 1.79 ± 0.70 pathogens with Methicillin-resistant Staphylococcus aureus being
days) the most common isolate. Acinetobacter baumanii was the major
Duration of hospital stay mean ± SD (in 24.2 ± 8.68 7.10 ± 2.29 gram-negative organism followed by Enterobacter and Escherichia
days) coli. The antibiogram of the major isolated pathogens have been
Case fatality rate (%) 87 (29) 37 (10)
mentioned in Table 3. The antibiotic sensitivity of organisms shows
emergence of resistant strains against Penicillin, Ciprofloxacin and
Erythromycin though retaining susceptibility to Vancomycin and
Risk factors Gentamycin. Gram-negative organisms have emerged resistant to
all the traditional drugs and are sensitive to new congener of antibi-
The general characteristic of the study and control group are otics.
depicted in Table 1. Both the groups were comparable with respect Table 4 shows the environmental surveillance report from the
to gender, mode of delivery, small for gestation age, apgar score NICU (low risk and high risk). The isolation percentage of the pre-
at 1 & 5 min less than 7, resuscitation requirement and malfor- dominant pathogen S. aureus in the blood culture reports from both
mation. On univariate analysis birth weight (<1500 g), gestation low and high risk NICU was noted per month for one year period
age (<37 weeks) and presence of hyaline membrane disease and its average was calculated. The average rate of isolation of S.
were significantly associated with higher incidence of HAI. Birth aureus in the high risk and low risk NICU was 15% and 9% respec-
weight (<1500 g) showed the highest association (OR 4.16; 95% CI tively. In the high risk NICU, the months of January, September and
2.57–6.74) followed by hyaline membrane disease presence (OR December had isolation rates of Staphylococcus species as 22%, 21%
3.42) and gestation age (<37 weeks) (OR 1.73). Male gender (p and 24% which exceeded average isolation rate. In the low risk
value 0.075) and apgar score at 5 min less than 7 (p value 0.086) NICU, the month of July had isolation rate of 13% above average.
were of borderline significance. On applying multivariable logis- The months of March, April, May and June had isolation rates of
tic regression to these risk factors, only male gender, gestation age S. aureus in the range from 9% to 11% but there was no growth in
(<37 weeks) and apgar score at 5 min less than 7 were identified as the surveillance report. Though it shows some vague association of
significant independent predictors. It underscored the importance HAI with the environmental surveillance report but its significance
of birth weight and hyaline membrane disease as independent value is yet to be known.

Table 2
Risk factors for HAIs.

Risk factors Odds ratio (OR) 95% confidence interval (CI) p Value Adjusted odds ratio (95% CI)

Male sex 1.36 0.96–1.91 0.075 1.21 (0.95–1.55)

Cesarean 1.29 0.90–1.80 0.157
Birth weight (<1500 g) 4.16 2.57–6.74 <0.0001
Gestational age (<37 weeks) 1.73 1.25–2.39 0.0009 1.55 (1.15–2.18)
Small for gestational age 0.86 0.60–1.25 0.450
Apgar score at 1 min < 7 0.89 0.59–1.34 0.596
Apgar score at 5 min < 7 2.45 0.85–7.04 0.086 1.92 (0.80–3.21)
Resuscitation required 0.95 0.65–1.35 0.764
Hyaline membrane disease 3.42 2.18–5.36 <0.0001
Malformation 0.80 0.33–1.98 0.646
Interventional risk factors
Peripheral vascular catheter use 15.39 10.22–23.18 <0.0001
Mechanical ventilator use 5.56 2.26–11.61 <0.0001
Peripheral vascular catheter duration <0.0001
Mechanical catheter duration <0.0001

Please cite this article in press as: Kumar S, et al. Healthcare associated infections in neonatal intensive care unit and its correlation
with environmental surveillance. J Infect Public Health (2017), http://dx.doi.org/10.1016/j.jiph.2017.08.005
G Model
JIPH-795; No. of Pages 5 ARTICLE IN PRESS
4 S. Kumar et al. / Journal of Infection and Public Health xxx (2017) xxx–xxx

Table 4

Vancomycin
Environmental surveillance report from low and high risk NICU.

Month Low risk NICU High risk NICU

100
100
January No growth Staphylococcus aureus—feeding
counter, phototherapy lights

Ciprofloxacin
February No growth No growth
May No growth No growth
June No growth No growth
July Staphylococcus aureus—suction No growth

95
65

65
70

90
tubing
Erythromycin August No growth No growth
September No growth Staphylococcus aureus – baby
cot Escherichia coli –
laryngoscope Acinetobacter
Others

baumanii – laryngoscope
85
38
50
October No growth No growth
November No growth No growth
Netilmycin

December No growth Staphylococcus aureus—radiant

warmer, environmental air
83
70
Amikacin

Discussion
85
87
50
Aminoglycosides

Developments in neonatal intensive care over the last few

Gentamycin

decades have increased the survival rate of neonates with low birth
weight and preterm infants, though HAI is still unsolved and sig-
nificant problem. In the present study male gender, gestational age
90
80
90

(<37 weeks) and apgar score at 5 min less than 7, PVC and MV use
along with their duration were identified independently as signif-
Ceftazidime

icant risk factors. Birth weight was dependent variable.

Babanzono et al. [10] studied risk factors in neonatal intensive
care unit using NICU surveillance data and showed that the inci-
95
50

dence of infection was significantly higher in the boys (OR 1.28; 95%
Piperacillin—tazobactum

CI 0.43–3.75). Orsi et al. [11] evaluated hospital-acquired infection

surveillance in 575 neonates admitted for >48 h NICU in which 76
developed a total of 100 HAIs (13.2%). Out of 76, 47(64%) were males
and 27 (36%) females. Similarly, this study suggests for increased
susceptibility of male babies to HAIs. However, prospective surveil-
lance study of HAIs done on 434 neonates by Gadallah et al. [3]
found female sex having increased risk for HAIs (RR 1.15; 95% CI
67
50

80

0.85–1.56) though insignificant (p value 0.351). Another study by

Piperacillin

Tavora et al. [12] showed gender as insignificant risk factor (p value

0.65). Secondly, this study demonstrated gestation age less than 37
Percentage susceptibility (%)

weeks had 1.35 times higher risk of acquisition as compared to

46
35
30

1.63 times for gestation age less than 38 weeks in the study done
Penicillin
␤-Lactam antibiotics

by Gadallah et al. [3]. However, Jeong et al demonstrated that ges-

tation age 32-36 weeks had increased risk though insignificant (RR
100
18
85
Generic name

1.21; p value 0.523) but 28–31 weeks (RR 3.05, p value <0.0001) and
≤27 weeks (RR 6.29; p value <0.0001) were significant risk factors.
Oxacillin

Finally, apgar score at 5 min less than 7 was shown to have signif-
100

icantly increased risk in resemblance to Jeong study [13] wherein

Antibiogram of Gram-positive and Gram-negative organism.

33
80

the apgar score between 4 to 7 & less than 4 at 5 min had 2.46 times
Methicillin susceptible Staphylococcus aureus (MSSA)

and 5.75 times increased risk (p value <0.0001). Other studies too
Methicillin resistant Staphylococcus aureus (MRSA)

have shown lower apgar score at 5 min as independent risk factor

in the acquisition of HAI [14].
In this study, PVC use was shown to have 15.39 times higher
risk of acquiring HAIs than MV use (RR5.56) mimicking the results
Coagulase negative Staphylococcus

of Tavora study [12] done in Brazil where PVC use OR 10.88

(95% CI 3.21–36.83) had higher association than MV use OR 3.42
(2.17–5.41). Though, Babanzono et al. [10] identified artificial ven-
Gram negative organism

Pseudomonas aeruginosa

tilation use as insignificant risk factor for MRSA infection (OR 0.97).
Gram positive organism

Acinetobacter baumanii

In contrast, Bolat et al. [4] in prospective study showed MV use asso-

ciated with 18.2 times (95% CI 12.4–26.7) higher risk than PVC (RR
Escherichia coli

7.1; 95% CI 4.3–12.5). This disparity was due to lesser mean ventila-
tor days. The mean duration of PVC and MV use in the study group
Organism

was 4.78 ± 2.44 days and 3.09 ± 0.98 days respectively. This showed
Table 3

similarity to the Pawa et al. study [15] in which the median duration
of PVC 210 (136–310) hours was greater than MV 60 (30–11) hours.

Please cite this article in press as: Kumar S, et al. Healthcare associated infections in neonatal intensive care unit and its correlation
with environmental surveillance. J Infect Public Health (2017), http://dx.doi.org/10.1016/j.jiph.2017.08.005
G Model
JIPH-795; No. of Pages 5 ARTICLE IN PRESS
S. Kumar et al. / Journal of Infection and Public Health xxx (2017) xxx–xxx
The case fatality rate in the study group was 29% as compared to
9% in the control group [3].
In this study, bloodstream infections (73%) accounting for the
majority with clinical sepsis (37%) and culture positive (36%) fol-
lowed by Pneumonia (12%), and meningitis (10%) as the 3rd most
common infection. Blood stream infection and Pneumonia were
the most common healthcare associated infection reported in the
literature with the following rates: 16%–78% for bloodstream infec-
tions and 10%–40% for pneumonia [16]. The percentage of blood
stream infections in this study is worrisome, since it causes greater
duration of hospital stay and increased mortality.
The microbiological profile in this study reveals predominance
of Gram-positive cocci isolates (67%) and Gram-negative bacilli
(33%). Among the bacteriological profile, S. aureus constituted the
majority (65%) out of which MRSA accounted for 40%. In study done
by Jeong et al. [13], the most frequently isolated organism was
S. aureus, followed by coagulase-negative Staphylococcus. Overall
Gram positives (60.2%) were more common than Gram-negatives
(31.8%). In similar study by Su et al. [17], the most common nosoco-
mial pathogens were coagulase-negative staphylococci (19%) and
S. aureus (14%).
The above average isolation rate of Staphylococcus aureus
matched with the monthly growth of similar organism in the
surveillance report. Correlating organisms from the healthcare
associated infection to the environmental surveillance may not
show cause-effect relationship. Environmental cultures may be one
of the predicting causative factors contributing to late onset sep-
sis but its statistical significance is yet to be proven. Molecular
typing will be required along with survival period of staphylococ-
cal species on different inanimate articles and collection of blood
samples should be restricted to that period. Our knowledge is still
lacking regarding answers to such questions which need more elab-
orate studies.
Conclusion
This study showed lower gestational age, male gender, apgar
score less than 7 at 5 min and interventional risk factors as indepen-
dent variables in the causation of healthcare associated infections.
These risk factors prolong duration of stay and add to increased
mortality rates. The bacteriological profile of HAI correlates with
the environmental surveillance report thus insisting for periodic
surveillance and thereby avoiding irrational antibiotic usage.
Conflict of interest
There are no conflict of interest.
Funding
None.
Please cite this article in press as: Kumar S, et al. Healthcare associated infections in neonatal intensive care unit and its correlation
with environmental surveillance. J Infect Public Health (2017), http://dx.doi.org/10.1016/j.jiph.2017.08.005
5
Acknowledgment
We would like to thank Dr. V. Sreenivas for his sincere effort in
statistical analysis. Harish Chellani and Manorma Deb conceptual-
ized the study and approved the final manuscript. Sanjay Kumar
collected and interpreted the data along with Dr. Binoy Shankar.
Sugandha Arya critically revised the manuscript for intellectual
contents.
References
[1] Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for noso-
comial infections. In: Olmsted RN, editor. APIC infection control and applied
epidemiology: principles and practice. St. Louis: Mosby; 1996. p. A-1–20.
[2] Craft A, Finer N. Nosocomial Coagulase-negative Staphylococcal catheter-
related sepsis in preterm infants: definition, diagnosis, prophylaxis and
prevention. J Perinatol 2001;21:186–92.
[3] Gadallah MAH, Fotouh AMA, Habil IS, Imam SS, Wassef G. Surveillance of
health care associated infections in a tertiary hospital neonatal intensive
care unit in Egypt: 1-year follow-up. Am J Infect Control 2014;42(November
(11)):1207–11.
[4] Bolat F, Uslu S, Bolat G, Comert S, Can E, Bulbul A, et al. Healthcare-
associated infections in a neonatal intensive care unit in Turkey. Indian Paediatr
2012;49(December (12)):951–7.
[5] Urzedo JE, Levenhagen MMD, Pedroso RS, Abdallah VOS, Sabino SS, Dolinger
Brito DV. Nosocomial infections in a neonatal intensive care unit during 16
years: 1997–2012. Rev Soc Bras Med Trop 2014;47(May–June (3)):321–6.
[6] Bokulich NA, Mills DA, Underwood MA. Surface microbes in the neonatal inten-
sive care unit: changes with routine cleaning and over time. J Clin Microbiol
2013:2617–24.
[7] Shankar MJ, Aggarwal R, Deorari AK, Paul VK. Sepsis in the Newborn. Indian J
Pediatr 2008;75(March):261–6.
[8] Koneman EW, Allen SD, Janda WM, Schreckenberger RC, Winn WC. Introduc-
tion to microbiology. Part II: guidelines for the collection transport, processing
analysis and reporting of culture from specific specimen sources. In: Kone-
man EW, Allen SD, Janda WM, Schreckenberger RC, Winn WC, editors. Color
atlas and textbook of diagnostic microbiology. 5th ed. Philadelphia: Lippincott
Raven; 1997. p. 121–70.
[9] Collee JG, Miles RS, Watt B. Tests for the identification of bacteria. In: Collee
JG, Fraser AG, Marmion BP, Simmons A, editors. Mackie and McCartney prac-
tical medical microbiology. 14th ed. New York: Churchill Livingstone; 1996. p.
131–49.
[10] Babanzono A, Kitajima H, Nishimaki S, Nakamura T, Shiga S, Hayakawa M, et al.
Risk factors for nosocomial infection in neonatal intensive care unit by the
Japanese infection surveillance (JANIS). Acta Med Okayama 2008;62(August
(4)):261–8.
[11] Orsi GB, d’Ettorre G, Panero A, Chiarini F, Vullo V, Venditti M. Hospital acquired
infection surveillance in a neonatal intensive care unit. Am J Infect Control
2009;37:201–3.
[12] Tavora AC, Castro AB, Militao MA, Girao JE, Ribeiro Kde C, Tavora LG. Risk factors
for nosocomial infection in a Brazilian neonatal intensive care unit. Braz J Infect
Dis 2008;12(Febraury (1)):75–9.
[13] Jeong IS, Jeong JS, Choi EO. Nosocomial infection in newborn intensive care unit,
South Korea. BMC Infect Dis 2006;6(June):103.
[14] Djordjevic ZM, Markovic-Denic L, Folic MM, Igrutinovic Z, Jankovic SM. Health
care acquired infections in neonatal intensive care units: risk factors and etiol-
ogy. Am J Infect Control 2015;43:86–8.
[15] Pawa AK, Ramji S, Prakash K, Thirupuram S. Neonatal nosocomial infection:
profile and risk factors. Indian Paediatr 1997;34(April):297–302.
[16] Tiemersma EW, Bronzwaer SLAM, Lyytikainen O, Degener JE, Schrijnemakers
P, Bruinsma N, et al. Methicillin resistant Staphylococccus aureus in Europe,
1999–2002. CDC Emerg Infect Dis 2004;10(9):1627–36.
[17] Su BH, Hsieh HY, Chiu HY, Lin HC, Lin HC. Nosocomial infection in a neona-
tal intensive care unit: a prospective study in Taiwan. Am J Infect Control
2007;35:190–5.