Three-year study of health care-associated infections in a Turkish pediatric ward

Introduction: Health care-associated infections (HCAIs) can cause an increase in morbidity, mortality and costs, especially in developing countries. As information on the epidemiology of HCAIs in pediatric patientsinTurkey is limited, we decided to study the annual incidence and antibiotic resistance patterns in our pediatric ward at Marmara University Hospital. Methodology: All hospitalized patients in the pediatric ward were assessed with regard to HCAIs betweenJanuary 1, 2008 and December 31, 2010. Data was prospectively collected according to standard protocols of the National Nosocomial Infections Surveillance System (NosoLINE). Results: A total of 16.5% of all hospitalized patients developed HCAIs in the three years studied. The most frequent HCAIs were urinary tract infections (UTI) (29.3%), bloodstream infections (27%) and pneumonias (21%). While the most frequent agent isolatedfrom UTI was Escherichia coli (26%), the most common agent in blood stream infections was Staphylococcus epidermidis (30.4%). Vancomycin resistance was found in 73.3% of all Enterococcus faecium strains. Extended-spectrum β-lactamase was detected in 58.3% of Klebsiella pneumoniae and E. coli isolates. Conclusions: Continual HCAI surveillance is important to determineits rate. Knowledge of the HCAI incidence can influence people’s use of broad-spectrum antibiotics and encourage antibiotic rotation. Moreover, the knowledge of HCAI incidence may support the infection control programmes, including education and isolation methods which ultimately may help to reducethe rate of the HCAIs.


Introduction
Health care-associated infections (HCAIs) are important complications in both adults and children that may lead to increased morbidity and mortality, prolonged hospital stay and increased costs [1].Incidence of HCAI varies according to age, service, underlying disease and other risk factors [2,3].In developed countries, the rates of HCAI among children are lower than among adults.For instance, in the United States 5%-10% ofadult patients hospitalized suffer HCAI while the rate is 1.5%-4% for children of ten years of age, and 7%-9% for infants younger than 1 year of age [4].This relationship between increased infection rate and younger age disappears in pediatric and neonatal intensive care units as the rate of HCAI reported for both is highdue to the increased severity of diseases and the need for more invasive procedures [5].In developing countries, the incidence of HCAI has been reported to be higher than in developed nations because of the high number of patients, limited number of staff, and insufficient compliance with infection control measures [6,7].
The most frequent HCAIs are bacteremia, urinary and respiratory tract infections [3,4,[8][9][10][11].The most common causative agents of HCAIs are staphylococci and Gram-negative organisms [4].However, information regarding the epidemiology of HCAIs in Turkish pediatric patients is limited.The aim of this study wasto assessthe epidemiology of HCAIs and species distribution as well as antimicrobial susceptibility of pathogens appearing in one of the Turkish University Hospitals.

Methodology
This study was performed in pediatric units consisting of 28 beds and 6 rooms and one pediatric intensive care unit at Marmara University Hospital (MUH).A total of 6 nurses and 5 doctors worked in this pediatric service during the day time.Continual active surveillance of HCAIswas performed by a nurse in charge for infection control.The MUH is located in themetropolitan Istanbul's Asian side.
This study included all patients hospitalized betweenJanuary1, 2008 and December 31, 2010 at MUH pediatric units.Data was prospectively collected according to standard protocols of the National Nosocomial Infections Surveillance System (NosoLINE).Centers for Disease Control and Prevention (CDC) criteria were used as standard definitions for HCAIs [12].HCAI was described as infection occurred 48 hours after admission or 10 days after discharge.Depending on symptoms, urine, cerebrospinal fluid (CSF), endotracheal aspirate, sputum, or wound specimens were obtained.Blood cultures were performed on all patients with suspected HCAI.
Blood cultures were performed using BACTEC peds plus/F bottles (BD Diagnostics, Sparks Glencoe, USA).Identifications were done using the VITEK2 (BioMérieux, Marcy l'Etoile, France).The Extendedspectrum β-lactamase (ESBL)was detected using the E-test, according to the manufacturer'sinstructions (AB Biodisk, Solna, Sweden).Susceptibility to non-βlactam antibiotics was evaluated by a disk diffusion method according to the Clinical and Laboratory Standards Institute (CLSI)criteria [13].E-test strips of vancomycin and teicoplanin were used to confirm resistance to glycopeptıdes according to the manufacturer'sinstructions (AB BIODISK, Solna, Sweden).For the interpretation of susceptibility results, the breakpoints of resistance set by the CLSI were used [13].Allinformation and culture results of patients with HCAI were collected by an infection control nurse.
Among agents isolated from HCAIs the frequency of methicillin resistance was 84.6% for S. epidermidis and seven out of nine Staphylococcus aureus strains.Among the gram-negative species obtained only from HCAIs, the 58.3% of K. pneumoniae and E.a coli isolateshad ESBL.C. albicans and non-albicans Candida strains accounted for 56% and 44% of HCAIs, respectively.Vancomycin and ampicillin resistance was found in 73.3% and 100% of all E. faecium strains, respectively.Infections due to vancomycin-resistant E. faecium strains led to an outbreak in the pediatric service.The susceptibility profile of all Acinetobacter baumanii and Pseudomonas aeruginosa isolates are shown in table 5.

Discussion
The overall HCAI rate of 16.5%detected by this study was consistent with results of previousstudies conducted in Turkey [7,14,15] which is higher than rates reported from Canada and US [16,17].Turkish HCAI ratesamong pediatric and adult patients in 2009, ranged from 1.3%-16% [7].In two previous studies that included children and adult patients at Pamukkale University Hospital and Marmara University Hospital,the HCAI rates were 3.5%-9.6%[14,15].The main reasons for these high rates were considered to be prolonged hospitalizations of patients with underlying chronic diseases, patients hospitalized in pediatric intensive care unit, and insufficient compliance with infection control measures.
In the US,gastrointestinal and respiratory infections and bacteremias are the most common HCAIs in the pediatric services [3].In patients hospitalized at Marmara University Hospital, the most common HCAI was represented byUTIs, (Table 1) probably due to frequent urinary tract catheterizations.E.coli was the most frequent agent in UTIs, and S. epidermidis was the most common agent in bacteremias.This wasin agreement with two separate studies regarding nosocomial bloodstream infections in United States hospitals [18,19].
More than 50% of E. coli strains and of K. pneumoniae isolates had ESBL in this current study, (Table 2) similar to a 2004 study conducted at Marmara University Hospital.In an international study that included Turkey it was reported that the 78% of K. pneumonia isolates produced ESBL [20].We considered these consistently high rates of resistance to be caused by insufficient compliance with infection control measures, stable patient profile of our hospital and unchanged physical conditions in the hospital [15].
P. aeruginosa frequently results in HCAIs and tends to develop multidrug resistance [21].In this study P. aeruginosa was the most common cause of pneumonia and the fifth most common cause of allHCAIs (Table 2).The resistance patterns in P. aeruginosa isolates varyin different areas of the hospital and varyin time,therefore continual surveillance and a timely provision of antibiograms mayhelp guide clinicians in selecting the empirical treatment.After the publication of another study from our group in 2004 [15], which showed that P. aeruginosa strains were susceptible to ceftazidime while being less susceptible to other antibiotics, clinicians began using broad-spectrum cephalosporins more frequently.This current study has documented a change to a reduced susceptibility to ceftazidime and anhigher susceptibility to amikacin and ciprofloxacin.
In 2004 we reported that 6 out of 13 isolates of E. faecium in the pediatric ward were resistant to vancomycin [15]; as a consequence antibiotic prescription practiceschanged to a more frequent use of broad-spectrum antibiotics.The change to broadspectrum antibiotics can be considered essential since the patients were seriously ill and may had been previously hospitalized.This current study has documented a 70% vancomycin resistance.
Although C.albicans was found to be the third most common agent causingurinary tract infection, C. albicans ranked sixth amongoverall causesof HCAIs in the present study (Table 2).This result maybe relatedto patients with hematologic or oncologic problems with a history of prior and/orlong-term hospitalization.

Conclusions
Compared to developed countries, the HCAI rate here reported was high, but it was in line with previously published reports from Turkey.As patients in our unit frequently need urinary catheterization, the most common HCAI was identified as urinary system infection.Antibiotic resistance rates are similar to other reports in Turkey.We hypothesize that thehigh rate of HCAIwith resistant bacteria reported in our study is caused by theexisting ward system, lack of infrastructure, and inability to implement infection control measures.Continual, active surveillance studies of hospital infections in developing countries, such as Turkey,is an essential component of infection control which maycontribute to improve patient care.

Table 2 .
Distribution of nosocomial pathogens and resistance status with respect to year of isolation These strains were obtained only from HCAIs

Table 3 .
Frequency of nosocomial pathogens causing UTI according to yearof isolation

Table 4 .
Frequency of nosocomial pathogens causing bloodstream infections according to yearof isolation

Table 5 .
Susceptibility profile of all Acinetobacter baumanii and Pseudomonas aeruginosa isolates