Access to this full-text is provided by Wiley.
Content available from BioMed Research International
This content is subject to copyright. Terms and conditions apply.
Research Article
Determining the Independent Risk Factors and Mortality Rate of
Nosocomial Infections in Pediatric Patients
Fesih Aktar,1Recep Tekin,2Ali GüneG,1Cevat Ülgen,3Elhan Tan,3Sabahattin ErtuLrul,3
Muhammet KöGker,4Hasan BalJk,3Duran Karabel,3and Ilyas YolbaG3
1Department of Pediatric Intensive Care Unit, Dicle University Medical School, 21010 Diyarbakir, Turkey
2Dicle University School of Medicine, Clinical Microbiology and Infectious Diseases, 21010 Diyarbakir, Turkey
3Department of Pediatrics, Dicle University Medical School, 21010 Diyarbakir, Turkey
4Department of Pediatric Infectious Diseases, Diyarbakir Children’s Hospital, Diyarbakir, Turkey
Correspondence should be addressed to Fesih Aktar; fesihaktar@yahoo.com
Received October ; Revised January ; Accepted January
Academic Editor: Benoˆ
ıt Stijlemans
Copyright © Fesih Aktar et al. is is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
e objective of this study was to determine the rate, independent risk factors, and outcomes of healthcare-associated infections
in pediatric patients. is study was performed between and in pediatric clinic and intensive care unit. patients and
control subjects were included in the study. Of patients with nosocomial infections (NIs), there were NIs episodes and
culture growths. e median age was . months. e median duration of hospital stay of the patients was . days. e most
frequent pathogens were Coagulase-negative Staphylococcus,Acinetobacter spp., Klebsiella spp., and Candida spp. Unconsciousness,
prolonged hospitalization, transfusion, mechanical ventilation, use of central venous catheter, enteral feeding via a nasogastric tube,
urinary catheter, and receiving carbapenems and glycopeptides were found to be signicantly higher in NIs patients. Multivariate
logistic regression analysis showed prolonged hospitalization, neutropenia, and use of central venous catheter and carbapenems
as the independent risk factors for NIs. In the univariate analysis, unconsciousness, mechanical ventilation, enteral feeding, use
of enteral feeding via a nasogastric tube, H2receptor blockers, and port and urinary catheter were signicantly associated with
mortality. In the multiple logistic regression analysis, only mechanical ventilation was found as an independent predictor of
mortality in patients with NIs.
1. Introduction
Healthcare-associated infections (HC-AIs) are becoming
an increasingly important health issue worldwide. HC-
AIs, which are generally observed among patients receiving
healthcare in hospital settings, are associated with increased
morbidity, mortality, and duration of hospitalization, as
well as increased treatment duration and costs. Patients
hospitalized in pediatric services and pediatric intensive care
units (PICUs) have a higher risk of nosocomial infections [].
Prolonged hospitalization, invasive interventions, congenital
malformations, and total parenteral nutrition are signicant
factors that increase the risk of nosocomial infections in
pediatric patients []. Knowing the risk factors that lead
to healthcare-associated infections in pediatric patients is
important for preventing nosocomial infections and reducing
associated mortality. ere are limited number of studies
investigating the risk factors and mortality of nosocomial
infections in pediatric services.
In this study, our aim was to determine the frequency,
type, and most common underlying factors of nosocomial
infections among patients hospitalized in the Pediatrics Clin-
ics and evaluate the eects of these infections on the duration
of hospitalization and the independent risk of nosocomial
infections.
2. Materials and Methods
e study was performed between September and April
in patients who were hospitalized, monitored, and
treated in the patient services and PICU of the Dicle Uni-
versity Children Hospital, excluding patients at the neonatal
Hindawi Publishing Corporation
BioMed Research International
Volume 2016, Article ID 7240864, 5 pages
http://dx.doi.org/10.1155/2016/7240864
BioMed Research International
service; all patients included in the study were hospitalized
formorethanhours.Atotalofpatientsandcontrol
subjects were included in the study. Surveillance of the noso-
comial infection was performed actively and prospectively by
a physician and a nurse, who checked the patient infections.
Patients were visited every day throughout their period of
hospitalization, and information regarding the patients was
recorded on the patient follow-up chart of the infection
control committee. Nosocomial infections were diagnosed
according to the Centers for Disease Control and Prevention
(CDC) criteria, which were based on the patients’ clinical and
physical examination ndings and laboratory results.
Blood, urine, sputum, cerebrospinal uid, wound site,
and catheter culture samples were obtained from every
patient assumed to have developed HAI; endotracheal aspi-
ratesampleswerealsoobtainedfrompatientsreceiving
mechanical ventilation. BACTEC Peds Plus/F (BD, Sparks
MD)® culture bottles were used for collecting blood samples.
e Automatized Phoenix culture system was used to identify
the microorganisms and their antibiotic sensitivities. e
reports prepared according to the standards of the Clinical
and Laboratory Institute, USA, were carefully evaluated.
Clinical and laboratory ndings were thoroughly evaluated
at the time of diagnosis. Laboratory results included positive
cultures, which were presumably obtained from sterile sites
(blood, CNS-uid, and pleural and peritoneal uids), as
well as peripheral leukocyte and platelet counts and C-
reactive protein (CRP) levels. New inltrations on radio-
logical images were also examined. e patients’ clinical
presentations included fever, pulmonary auscultation nd-
ings, and hypotension. e clinical, radiological, laboratory,
and culture results of HAI patients were recorded on the
standard forms of the NNIS-system, and all relevant data
were recorded daily onto computers. HAI diagnosis was
established according to the criteria set by the CDC.
Peripheral blood culture samples and blood samples
obtained via sterile catheters were collected from patients
monitored at pediatric services and PICU who were sus-
pected of having bacteremia. ese samples were assayed
with the BacT/Alert (Biomerieux, France) automated blood
culture system; positive-testing samples were subjected to
further evaluation. ree to four cm sections from the tip
of central venous catheters were seeded semi-quantitatively
under sterile conditions according to the Maki method into
blood agar and EMB agar. Samples of urine and tracheal
aspirate were collected under sterile conditions and seeded
into sheep blood agar and EMB plates. e diagnosis of
urinary tract infection (UTI) was done by the detection of
bacteria in the urine sample drawn from the catheter with
at least . CFU/mL of a single or two dierent species.
A fully automated identication and antibiogram device was
used for the bacterial identication and antibiogram.
Patients without nosocomial infections who were hos-
pitalized at the pediatric clinic throughout study period
were recruited as controls. Control subjects were specically
selected from patients hospitalized at the same clinic and with
similar ages.
Prolonged hospitalization is dened as hospitalization
equalorlongerthantwoweeks.Neutropeniaisdenedas
anabsoluteneutrophilcountoflessthan/𝜇Lorlessthan
/𝜇L with an anticipated decline to less than /𝜇Linthe
next -hour period.
Ethical approval for the study was obtained from the
Noninterventional Clinical Research Ethics Committee of
Dicle University Medical Faculty.
3. Statistical Analysis
Data were analyzed using the SPSS . (SPSS Inc., Chicago,
IL, USA) statistical soware. Visual (histogram) and ana-
lytical methods (Kolmogorov-Smirnov test) were used to
determine whether variables exhibited normal distribu-
tion. Variables with normal distribution were expressed as
mean plus/minus standard deviation, while not normally
distributed data were expressed as median (minimum-
maximum) values. Independent groups were compared by
Student’s 𝑡-test. Categorical data were compared using the
Chi-square test. Correlations were determined using Pear-
son’s or Spearman’s correlation analyses. To identify inde-
pendent factors that may inuence disposition, multivariable
logistic regression analyses (polytomous responses) were
performed to calculate the odds ratio and corresponding
% condence intervals. 𝑝value <. was considered
statistically signicant.
4. Results
e primary outcome of the study was the development of
HC-AI. ere were episodes of nosocomial infections in
patients. Among these cases with hospital infections,
were female (%) and were male (%). Of the control
subjects, were male (.%) and were female (.%).
e median age and interquartile ranges (th and th per-
centile) of the patients and control group were . (.–.)
months and . (.–.) months, respectively. ere were
no signicant dierences in the mean age and gender distri-
bution between the study and the control groups (𝑝 > 0.05).
e median duration of hospital stay and interquartile ranges
(th and th percentile) of the patients and control group
were . (.–.) days and . (.–.) days, respec-
tively. Ventilator utilization ratio was . and central venous
catheter utilization ratio was .. e VAP rate was . per
ventilator days, and the catheter associated bloodstream
infection rate was . per catheter days. When patients
with HC-AI were evaluated with respect to the hospitalized
departments, it was determined that (.%) were moni-
tored in PICU, (.%) were in the pediatric hematology
clinic, and (.%) were monitored in other clinics. In the
control group, (%) subjects were monitored at the PICU,
(%) were monitored at the pediatric hematology clinic,
and (%) were monitored at other clinics. No signicant
dierence was found in clinical distribution between patients
with and without NIs (𝑝 > 0.05). e most frequent types of
infections were bloodstream infections, ventilator-associated
pneumonia, catheter associated bloodstream infections, and
urinary tract infections (Table ). e most commonly iso-
lated microorganisms were Coagulase-negative Staphylococ-
cus (%), Klebsiella spp. (.%), Acinetobacter spp. (.%),
BioMed Research International
T : Distribution of nosocomial infections according to the
infection sites.
Type of nosocomial infection Number (%)
Bloodstream infection ()
Ventilator-associated pneumonia ()
Catheter associated bloodstream infection ()
Urinary tract infection ()
Urinary catheter related infection ()
Shunt infection ()
Pneumonia ()
Gastrointestinal infection ()
Skin and so tissue infection ()
Nosocomial meningitis ()
Surgical site infection ()
Decubitus ulcer ()
Total 100 (100)
and Candida spp. (.%). Methicillin-Resistant S. aureus
(MRSA) was found less than % of all NIs isolated microor-
ganisms. And extended spectrum betalactamases (ESBL)
ratio was observed with percentage.
Univariate analysis showed that unconsciousness, pro-
longed hospitalization ( weeks), transfusion, mechanical
ventilation, enteral feeding via a nasogastric tube, use of
central venous catheter, urinary catheter, carbapenems, and
glycopeptides were signicantly associated with nosocomial
infections (Table ). Variables with 𝑝value <. were
selected for the logistic regression model. Nine variables (pro-
longed hospitalization, neutropenia, transfusion, mechanical
ventilation, use of central venous catheter, enteral feeding
via a nasogastric tube, carbapenems, aminoglycosides, and
glycopeptides) were selected among other similar variables
basedonclinicaljudgement.Inmultiplelogisticregression
analysis, prolonged hospitalization, neutropenia, use of cen-
tral venous catheter, and carbapenems were found to be
independent risk factors for nosocomial infections among
patients with HC-AIs (𝑅square = .) (Table ).
In the univariate analysis, unconsciousness, mechanical
ventilation, enteral feeding, use of enteral feeding via a
nasogastric tube, H2receptor blockers, and port and urinary
catheter were found to be signicantly associated with mor-
tality. Variables with 𝑝value <. were selected for logistic
regression model. Due to the small sample size (𝑛=26),
we selected three variables (mechanical ventilation, use of
H2receptor blockers, and use of urinary catheter) among
similarvariablesbasedonclinicaljudgement.Inthemultiple
logistic regression analysis, only mechanical ventilation (OR:
.,%CI,.–,𝑝 = 0.003,𝑅square: .) was
determined to be an independent predictor of mortality in
patients with nosocomial infection (Table ).
5. Discussion
Healthcare-associated infections (HC-AIs) are the most
important causes of morbidity and mortality in the pediatric
clinic patients. Recently, with the improvement in the like-
lihood of survival of pediatric patients who have been
under risk, HC-AIs are becoming an increasingly important
problem in PICUs []. In pediatric patients, HC-AIs are a
determining factor of morbidity and mortality; therefore, it
is important to know which factors are associated with the
development of infections in children, a population on which
there are only limited studies. Determining the risk factors
of HC-AIs may help decrease the incidence of infections
and reduce healthcare costs. On the other hand, long-term
monitoring, invasive interventions, total parenteral nutrition,
and the use of high-spectrum antibiotics are factors that
increase the risk of infection among patients who are treated
and monitored in pediatric clinics, especially in PICUs and
pediatric hematology clinics [–]. In parallel with the nd-
ings of earlier studies, the frequency of nosocomial infections
was highest in our study among PICU patients.
e most frequently observed infections in our study
were bloodstream infections, ventilator-associated pneumo-
nia, and urinary tract infections. e ratio of nosocomial
infections varies between % and % depending on the fol-
lowing risk factors: the number of patients being in the care,
frequency of invasive interventions, number of skilled health
personnel, medical equipment and infrastructure, and the
types of medical treatment oered []. A systematic review
performed by Balaban et al. in Turkey reported results similar
to those of our study []. Nosocomial infections increase
not only morbidity and mortality rates, but also responsible
prolongation of hospitalization and increase healthcare costs
[–]. e outcome of nosocomial infections depends on the
causative pathogen; for example, if bloodstream infections are
due to Coagulase-negative Staphylococci,theyareassociated
with increased mortality. In addition, the sensitivity of the
pathogen to the empirical antimicrobial treatment may play
an important role [, ].
Microorganisms which are responsible for nosocomial
infections vary not only from year to year but also between
countries []. S. aureus was a frequently identied as
causative agent in previous years, while in the following years
and today, Gram-negative microorganisms and Coagulase-
negative Staphylococci are the most frequently identied
causative pathogens []. In our study, MRSA represented
a very low percentage (less than %) of causative agents.
An earlier study performed by Erdem et al. indicated that
the incidence of S. aureus infections was declining rapidly
in Turkish intensive care units (ICUs), with potential impli-
cations on empirical treatment strategies []. Acinetobacter
with multiple drug resistance also represent a serious problem
recently.isproblemismainlyassociatedwithlong-term
hospitalization, use of inappropriate antibiotics, and failure
to follow adequately rules of infection control. In our study,
prolonged use of wide-spectrum antibiotics was determined
to be a risk factor for the HC-AIs. Similarly in the study
of Tekin et al., prolonged use of wide-spectrum antibiotics
was determined to be a risk factor for multidrug-resistant
Acinetobacter []. Acinetobacter infections are associated
with increase in mortality rates and prolongation of hospital
stay. erefore, unnecessary antibiotic use and prolonged
BioMed Research International
T : Comparison of risk factors between infected and uninfected groups.
Risk factors
Uninfected
patients
(𝑛=86)
Infected
patients
(𝑛=86)
𝑝
Sex %
Female . . .
Male . .
Age (month)∗. (.–.) . (.–.) .
Intensivecarestay% . . .
Length of hospital stay (days)∗. (.–.) . (.–.) <0.001
Unconsciousness % . . <0.001
Prolonged hospitalization % . . <0.001
Use of central venous catheter % . . <0.001
Enteral feeding via a nasogastric tube % . . <0.001
Tra n sfusi o n % . . <0.001
Parenteral nutrition % . . 0.036
Use of H2receptor blockers % . . 0.016
Mechanical ventilation % . . <0.001
Use of urinary catheter % . . <0.001
Neutropenia % . . 0.003
Use of port % . . 0.036
Use of third-generation cephalosporins % . . 0.049
Use of carbapenems % . . <0.001
Use of steroid % . . .
Use of aminoglycosides % . . 0.001
Use of glycopeptides % . . <0.001
∗Median and interquartile ranges (th and th percentile).
T : Multivariate logistic regression analysis in prediction of independent risk factors for hospital infections.
Unadjusted Adjusted
OR (% CI) 𝑝OR (% CI) 𝑝
Prolonged hospitalization . (.–.) <. . (.–.) 0.002
Neutropenia . (.–.) . . (.–.) 0.027
Use of central venous catheter . (.–.) <. . (.–.) 0.043
Transfusion . (.–.) <. . (.–.) .
Enteral feeding via a nasogastric tube . (.–.) <. . (.–.) .
Mechanical ventilation . (.–.) <. . (.–.) .
Use of carbapenems . (.–.) <. . (.–.) 0.034
Use of glycopeptides . (.–.) <. . (.–.) .
Use of aminoglycosides . (.–.) . . (.–.) .
R square = 0.544.
hospitalization should be avoided, especially in patients who
weretreatedandmonitoredinPICUsandhematologyclinics.
We observed a mortality rate of .% in present study,
which is comparatively higher than the ratio reported by
another study that identied a mortality rate of .% []. We
determined that unconsciousness, mechanical ventilation,
enteral feeding, use of enteral feeding via a nasogastric
tube, H2receptor blockers, and existence of port and uri-
nary catheters are the risk factors that lead to increased
mortality rate among pediatric patients. Mechanical ven-
tilation was determined to be an independent predictor
of mortality in patients with nosocomial infections. In a
study of Hacımustafao˘
glu et al., endotracheal intubation,
urinary catheter, and male gender were determined to be
independent risk factors for mortality []. e presence of
independent risk factor in our study group and the predom-
inance of ventilator-associated pneumonia were associated
with higher mortality rate. Due to the relatively limited
BioMed Research International
T : Multivariate logistic regression analysis in prediction of independent risk factors for mortality.
Unadjusted Adjusted
OR (% CI) 𝑝OR (% CI) 𝑝
Unconsciousness . (.–.) <.
Mechanical ventilation . (.–.) <. . (.–) 0.003
Use of enteral feeding via a nasogastric tube . (.–.) <. . (.–.) .
Use of H2receptor blockers . (.–.) <. . (.–.) .
Enteral feeding . (.–.) <.
Use of port . (.–.) <.
Use of urinary catheter . (.–.) <.
number of patients in our study, we suggest that further
comprehensive prospective studies need to be conducted in
order to better determine the factors aecting mortality.
Determining the variable factors associated with noso-
comial infections and taking the necessary measures against
them will help reduce morbidity and mortality rates. It is
possible to reduce the incidence of HC-AIs through a number
of signicant strategies and practices. ese should include
measures such as appropriate hand washing, before and aer
every contact with patients; periodic training of the health
workers; ensuring hygiene in the clinical environment; devel-
oping principles on the use of central venous catheters; limit-
ingtheuseofinvasivecatheters;andrationaluseofantibiotics
for treatment and prophylaxis of infections. In addition,
conducting regular surveillance activities in hospitals and
reviewing the associated surveillance data, determining the
potential causative infectious agents in ICUs, and detecting
resistance of infectious agents to antibiotics will contribute to
the management of nosocomial infections [, ].
In conclusion, nosocomial infections represent a partic-
ularly important issue in pediatric clinics and PICUs. Close
monitoring may decrease the rates of healthcare infections
and mortality. Measures for controlling infections, such as
ensuring compliance to hand hygiene practices, reducing
the duration of hospital stay for patients, and preventing
improper use of antibiotics, all will contribute to reducing the
incidence of nosocomial infections and related mortality.
Conflict of Interests
e authors declare that there is no conict of interests
regarding the publication of this paper.
References
[] J. Raymond and Y. Aujard, “Nosocomial infections in pediatric
patients: a European, multicenter prospective study,” Infection
Control and Hospital Epidemiology,vol.,no.,pp.–,
.
[] ˙
I. Balaban, G. Tanır, ¨
O. M. Timur et al., “Nosocomial infections
in the general pediatric wards of a hospital in Turkey,” Japanese
Journal of Infectious Diseases,vol.,no.,pp.–,.
[]S.S.Cavalcante,E.Mota,L.R.Silva,L.F.Teixeira,andL.B.
Cavalcante, “Risk factors for developing nosocomial infections
among pediatric patients,” Pediatric Infectious Disease Journal,
vol. , no. , pp. –, .
[] E. Kepenekli, A. Soysal, N. Yalindag-Ozturk et al., “Healthcare-
associated infections in pediatric intensive care units in Turkey:
a national point-prevalence survey,” Japanese Journal of Infec-
tious Diseases,vol.,no.,pp.–,.
[] M. Hacımustafao˘
glu, S. C¸ elebi, E. Tuncer, G. ¨
Ozkaya, D. C¸akır,
and S. Elmas Bozdemir, “Nosocomial infection incidence in
pediatric clinic and pediatric intensive care unit,” Journal of
Pediatric Infection,vol.,no.,pp.–,.
[] M.Urrea,M.Pons,M.Serra,C.Latorre,andA.Palomeque,
“Prospective incidence study ofnos ocomial infectionsin a pedi-
atric intensive care unit,” Pediatric Infectious Disease Journal,
vol. , no. , pp. –, .
[]M.K.Lee,C.S.Chiu,V.C.Chow,R.K.Lam,andR.W.
Lai, “Prevalence of hospital infection and antibiotic use at a
university medical center in Hong Kong,” Journal of Hospital
Infection,vol.,no.,pp.–,.
[] K. Rutledge-Taylor, A. Matlow, D. Gravel et al., “A point preva-
lence survey of health care-associated infections in Canadian
pediatric inpatients,” American Journal of Infection Control,vol.
, no. , pp. –, .
[] M.L.Abramczyk,W.B.Carvalho,E.S.Carvalho,andE.A.S.
Medeiros, “Nosocomial infection in a pediatric intensive care
unit in a developing country,” Brazilian Journal of Infectious
Diseases,vol.,no.,pp.–,.
[] K. A. Bryant and C. R. Woods, “Healthcare-acquired infections
due to gram-positive bacteria,” Pediatric Infectious Disease
Journal,vol.,no.,pp.–,.
[] H. Erdem, M. Dizbay, S. Karabey et al., “Withdrawal of
Staphylococcus aureus from intensive care units in Turkey,”
American Journal of Infection Control, vol. , no. , pp. –
, .
[] R. Tekin, T. Dal, F. Bozkurt et al., “Risk factors for nosocomial
burn wound infection caused by multidrug resistant Acineto-
bacter baumannii,” Journal of Burn Care and Research,vol.,
no. , pp. e–e, .
[] M. J. G. de Mello, M. D. F. P. M. de Albuquerque, H. R. Lacerda,
W. V. de S ouza, J. B. Correia, and M. C. A. de Britto, “Risk factors
for healthcare-associated infection in pediatric intensive care
units: a systematic review,” Cadernos de Saude Publica,vol.,
no.,pp.S–S,.
[] R. Tekin, T. Dal, H. Pirinccioglu, and S. E. Oygucu, “A -year
surveillance of device-associated nosocomial infections in a
neonatal intensive care unit,” Pediatrics and Neonatology,vol.
,no.,pp.–,.
Available via license: CC BY
Content may be subject to copyright.
Content uploaded by Recep Tekin
Author content
All content in this area was uploaded by Recep Tekin on Mar 14, 2016
Content may be subject to copyright.
