WITHDRAWN: Risk Factors and Interventions for Ventilator-Associated Pneumonia in Pediatric Patients

Journal of Pediatric Nursing (2010) xx, xxx–xxx

Risk Factors and Interventions for Ventilator-AssociatedPneumonia in Pediatric Patients1

Jill Morinec BSN, RN, CPNa, Jacalyn Iacaboni BSN, RN, CPNb, Molly McNett PhD, RNb,⁎

aPediatric Intensive Care Unit, Department of Nursing, MetroHealth Medical Center, Cleveland, OHbPediatric Intensive Care Unit and Adult Trauma/Critical Care, Department of Nursing, MetroHealth Medical Center,Cleveland, OH

C

0d

Key words:Ventilator associatedpneumonia;

Pediatric;Intensive care unit

Ventilator-associated pneumonia (VAP) is a leading nosocomial infection in pediatrics. Little researchhas investigated the risk factors or effectiveness of interventions for pediatric VAP prevention. Thepurposes of this study were to identify the risk factors associated with VAP in pediatric patients anddescribe current VAP prevention practices. Data were gathered retrospectively on ventilated patientsadmitted to the pediatric intensive care unit over 12 months. No variables were found to be predictive ofVAP. Review of practices indicates that better documentation is needed of all interventions. Findingsprovide information to guide the implementation of VAP bundles. Implementation should focus onadequate documentation of VAP prevention efforts.© 2010 Elsevier Inc. All rights reserved.

VENTILATOR-ASSOCIATED PNEUMONIA (VAP) isthe second most common nosocomial infection in pediatricintensive care units (PICUs), accounting for 20% of allnosocomial infections in the pediatric population (Tablan,Anderson, Besser, Bridges, & Hajjeh, 2003). VAP adverselyaffects patient outcomes and results in substantial morbidityand mortality, which in turn prolongs hospital length of stay(LOS) and increases medical cost (Alumuneef, Memish,Balkhy, Alalem, & Abutaleb, 2004). The Centers for DiseaseControl has recommended using a series of evidence-basedinterventions (i.e., a VAP bundle) to reduce VAP amongventilated adult patients. Implementation of these VAPbundles using an all-or-nothing approach has been shown toaddress the risk factors associated with VAP and reduceVAP considerably in the adult population (Fulbrook &Mooney, 2003). However, little research has investigated theeffects of these VAP bundles among ventilated pediatricpatients. To investigate the effects of VAP bundles amongpediatric patients in subsequent studies, baseline data are

1 This study was conducted at the MetroHealth Medical Center,leveland, OH.⁎ Corresponding author: Molly McNett, PhD, RN.E-mail address: [email protected] (M. McNett).

882-5963/$ – see front matter © 2010 Elsevier Inc. All rights reserved.oi:10.1016/j.pedn.2010.07.010

needed (a) to identify if pediatric risk factors for VAP aresimilar to risk factors in the adult population and (b) todocument current VAP prevention interventions performedamong ventilated pediatric patients. Thus, the aims of thisstudy were to identify the presence of risk factors associatedwith VAP among ventilated PICU patients and to describebaseline pediatric VAP prevention practices.

Background

VAP: Definition and Outcomes

VAP is defined as a nosocomial pneumonia in patientsreceiving mechanical ventilation via an endotracheal tube(ETT) that develops 48 hours or more after the initiation ofventilation (Curley et al., 2006). VAP results in increasedmorbidity and mortality rates, LOS, and health care costs(Alumuneef et al., 2004; Curley et al., 2006; Powers, 2006).In one study, Alumuneef et al. (2004) found that pediatricpatients with VAP had a mean duration of mechanicalventilation of 21.4 days compared with only 10.4 days ofmechanical ventilation in patients who did not develop VAP.

2 J. Morinec et al.

A separate study reported a 30.5% VAP rate, where pediatricpatients had a 21% mortality rate, and 47% of these deathswere found to be secondary to VAP (Patra, Jayashree,Singhi, Ray, & Saxena, 2007). Other studies have reportedVAP rates ranging from 4% to 44% (Alumuneef et al., 2004;Bigham et al., 2009; Hsieh, Wu, Lin, Chang, &Wong, 2010;Taira et al., 2009; Yildizdas, Yapicioglu, & Yilmaz, 2002).From these studies, it is evident that interventions to preventVAP are necessary to improve patient outcomes anddecrease health care costs.

Risk Factors for VAP

Studies conducted among ventilated adult patientsidentify risk factors associated with increased VAP rates.Modifiable risk factors include mechanical ventilation longerthan 48 hours, high gastric residuals, supine position, use ofantibiotics, frequent changes in ventilator circuits, presenceof nasogastric tubes (NGTs), and use of paralytic agents orcontinuous sedation (American Thoracic Society [ATS] &Infectious Disease Society of America [IDSA], 2005;Bonten, Kollef, & Hall, 2004; Collard, Saint, & Matthay,2003). Nonmodifiable risk factors include patient age,gender (male), coma, illness severity, trauma, preexistingpulmonary disease, and altered level of consciousness (ATS& IDSA, 2005; Bonten et al., 2004; Collard et al., 2003).Modifiable risk factors for VAP are amenable to nursing,medical, and respiratory interventions, many of which areincluded in adult VAP bundles. However, little research hasexamined to what extent these modifiable and nonmodifiablerisk factors identified in adults influence VAP rates amongventilated pediatric patients.

Researchers have just begun to examine the risk factorsassociated with VAP in pediatric patients. Studies providepreliminary information on risk factors that may be associatedwith the development of pediatric VAP. These include lengthof mechanical ventilation, use of opiates, sustained neuro-muscular blockade, presence of enteral nutrition, priorantibiotic therapy, endotracheal suctioning, reintubation,gastroesophageal reflux, subglottal/tracheal stenosis, agegreater than 10 years, and trauma (Abdel-Gaward, El-Hodhod,Ibrahim, & Michael, 2009; Alumuneef et al., 2004; Bighamet al., 2009; daSilva, Neto, deAguiar, Lopes, & deCarvalho,2010; Hsieh et al., 2010; Patra et al., 2007; Taira et al., 2009;Yuan, Chen, &Yu, 2007). Although some of these risk factorsmirror those identified in the adult population as contributingto VAP, many findings were based on small sample sizes, andit has been concluded that additional research is necessary toprovide a basis upon which to guide clinical practice inpediatrics. Specifically, evidence is needed to better identifyand confirm these risk factors.

Interventions for VAP

Numerous interventions have been studied and shown todecrease the incidence of VAP in the adult intensive care

unit setting (Curley et al., 2006), as they directly impactmodifiable risk factors for VAP. Specifically, use of aseries of interventions (i.e., a VAP bundle) using an all-or-nothing approach is recommended (Curley et al., 2006).The Institute for Healthcare Improvement 100,000 LivesCampaign recommended that the adult VAP bundle includehead of the bed (HOB) elevated 30° to 45°, daily sedationvacations, peptic ulcer disease prophylaxis, and deep veinthrombosis (DVT) prophylaxis (Curley et al., 2006). Evans(2005) implemented this VAP bundle to standardize care inthe adult medical intensive care unit of a 700-bed tertiarycare hospital. Use of this bundle showed improvement inVAP rates within 4 months and achieved an 88% reductionin VAP over 2.6 years, a 1.4-day decrease in monthly LOS,and a 3% reduction in mortality (Evans, 2005). Othersmaller studies implementing the VAP bundle in an all-or-nothing approach have yielded similar results, decreasingVAP rates and hospital costs (Resnar et al., 2005;Youngquist et al., 2007).

To date, only one published article details the effective-ness of a VAP bundle in pediatrics. A quality improvementinitiative conducted by Bigham et al. (2009) demonstrated areduction in VAP rates from 5.6 to 0.3 infections per 1,000ventilator days after the implementation of a VAP bundle(Bigham et al., 2009). A multidisciplinary, multicomponentstrategy was used to implement this VAP bundle inpediatrics, and data were gathered before and afterimplementation. The VAP bundle included many interven-tions seen in adult VAP prevention bundles, such as handhygiene, HOB elevation, routine oral care, and changingventilator circuits. The pediatric bundle also includedutilization of heated wire ventilator circuits to decreaseaspiration risk with repositioning. Unlike adult bundles, thispediatric VAP bundle did not include a component forgastrointestinal (GI) ulcer prevention. Implementation of thepediatric VAP bundle significantly decreased VAP rates,which was associated with PICU LOS, ventilator days, andmortality rates (Bigham et al., 2009).

Although implementation of VAP bundles has shown tobe highly effective in the adult population, additionalresearch is needed to substantiate the findings of Bigham etal. (2009) and to identify specifically what componentsshould be included in a pediatric VAP bundle. It is assumedthat many low-risk practices proven in the adult populationto prevent VAP may frequently already be in use in thepediatric critical care setting (Curley et al., 2006).Interventions such as HOB elevation, extubation readinesstesting, and peptic ulcer prophylaxis are likely used in thepediatric setting routinely (Curley et al., 2006), yet there areno data to document that these practices are consistentlyimplemented, their effects on VAP rates, and the extent towhich other risk factors associated with VAP are presentamong ventilated pediatric patients. More information isalso needed to determine the applicability of VAP bundles(Foglia, Meier, & Elward, 2007; Lachman & Yuen, 2009;Turton, 2008).

3VAP Risk Factors and Interventions

Conceptual Framework

The review of the literature presented here provides therationale for the choice of study variables and conceptualframework depicted in Figure 1. Variables that have beenlinked to VAP are influenced by (a) nursing interventions,(b) respiratory interventions, and (c) patient variables(modifiable and nonmodifiable risk factors for VAP). Theindependent variables for this study are therefore organizedand presented in the model using these three categories. Theprimary outcome variable for the study is the development ofVAP. Secondary outcome measures include PICU andhospital LOS.

Methods

This study was reviewed and approved by the hospitalinstitutional review board. A retrospective cohort study design

Figure 1 Conceptual mode

was used to identify risk factors and interventions for VAP inthe PICU. Medical records from all mechanically ventilatedpediatric patients admitted to the PICU at a Level I traumacenter over a 12-month period were reviewed. To be included,patients had to be admitted over the 12-month period andventilated for at least 48 hours. Records were excluded ifpatients had immunological disorders or a known underlyinglung infection or weremechanically ventilated for less than 48hours. Charts of patients meeting inclusion criteria werereviewed concurrently by two researchers using a standard-ized chart review instrument developed by the principalinvestigator for use in this study. Random audits were done bya third study investigator to verify the reliability and validityof the data abstracted from the first two investigators.

Data on the following study variables were gathered: (a)patient variables (age, gender, ethnicity, primary diagnosis,medical history, admitting service, oral ETT size, route ofintubation, sedation/paralytic medication, Glasgow ComaScale [GCS] score, Riker score, number of ventilator days,presence of NGT, GI feedings, GI ulcer prophylaxis, and

l of the study variables.

Table 1 Summary of Sample Demographic Data

Variable n %

Age0–28 days 2 5.329 days–12 months 16 42.112 months–11 years 12 31.612–18 years 8 21.1

GenderMale 24 63.2Female 14 36.8EthnicityCaucasian 17 44.7African American 16 42.1

4 J. Morinec et al.

DVT prophylaxis), (b) nursing interventions for VAP (oralcare, patient repositioning, HOB elevation, suctioningmethod, and frequency of suctioning), and (c) respiratoryinterventions for VAP (cuff pressure check, ventilator circuitchange, and ventilator condensation removal).

The primary outcome variable in this study was thepresence of VAP. VAP was defined as at least 48 hours ofmechanical ventilation, a change in the chest x-ray (showinginfiltrates or worsening of the x-ray), and at least two of thefollowing: (a) fever of at least 38.5°C, (b) change in whiteblood cell count (serum laboratory results show b4,000 orN10,000 cells/μl), or (3) positive tracheal aspirate culture.Secondary outcome variables were PICU and hospital LOS.

Other 5 7.9Primary diagnosisRespiratory distress 8 21.1Closed head injury 6 15.8Respiratory syncytial virus 6 15.8Other 11 28.9Medical historyNone 12 31.6Extracranial injuries 9 23.7Premature 7 18.4Route intubationOral 38 100Nasal 0 0

Data Analysis

All data were analyzed using the Statistical Package forthe Social Sciences software, Version 15.0. Descriptivestatistics, including means, frequencies, and standard devia-tions were first performed. Correlational and logisticregression analyses were then conducted to identify therelationships among study variables and to identify whichrisk factors were associated with VAP.

ETT typeUncuffed 16 42.1Cuffed 22 57.9

Table 2 VAP Patient Factors

Variable Percentage of Time Ordered

Continuous sedation 76.6Intermittent sedation 89.0Paralytic medicationYes 8.4No 91.6

NGTYes 93.7No 5.9

Continuous GI feedsYes 45.2No 54.8

GI decompressionYes 53.1No 46.9

GI ulcer prophylaxisYes 69.0No 27.2

SCDsYes 10.5No 89.5

Results

During the study period, there were 128 pediatric patientsadmitted to the PICU requiring mechanical ventilation. Ofthis number, 38 met the inclusion criteria. Several patientswere excluded because they were intubated for less than therequired 48 hours, had an underlying lung infection, orunderwent tracheostomy. Nine patients fit the VAP criteriaas outlined previously, whereas the resulting 29 of 38patients included in the study did not develop VAP.

Table 1 presents a summary of the descriptive data of thestudy sample. Most patients (42.1%) were between 1 and 12months of age and were male (63.2%). Approximately halfof the study population were Caucasian (44.7%) or AfricanAmerican (42.1%). Many had a primary diagnosis ofrespiratory distress (21.1%), and some (31.6%) had nohistory of previous medical problems. All participants(100%) were orally intubated, and slightly more than half(57.9%) had cuffed ETTs. The mean GCS score for patientswas 7.85 (SD = 2.006), and the mean Riker score was 3.59(SD = 8.941).

The frequencies of patient-level VAP risk factors aredisplayed in Table 2. All patients were receiving some typeof sedation and pain medication. Seventy-six percent ofpatients received continuous pain/sedation medications suchas propofol, fentanyl, and/or versed drips, whereas intermit-tent administration of pain/sedation medications was presentin 89% of patients. Only 8.4% of patients received paralyticmedications, such as vecuronium. Most patients (93.7%) had

NGTs in place. However, only 45.2% received continuousGI feedings, and about half (53.1%) had these NGTs to lowsuction (GI decompression). Most patients (69%) had GI

5VAP Risk Factors and Interventions

ulcer prophylaxis medications ordered. Few patients (10.5%)had sequential compression devices (SCDs) ordered forDVT prophylaxis.

Comparison of Risk Factors Between Groups

Chi-square tests for independence were performed toexplore if there were significant differences among categor-ical level risk factors when comparing theVAP (n = 9) and theno VAP (n = 29) groups. The following variables wereincluded in these analyses: ETT size, presence of GI feeds,presence of continuous sedation, ETT type (cuffed/uncuffed),primary diagnosis, gender, presence of NGT, GI prophylaxis,and presence of SCDs. Among these study variables, nonereached statistical significance between groups.

Mann–Whitney U analyses were then performed todetermine if there were significant differences betweengroups (VAP/non-VAP) for hospital and PICU LOS. Table 3displays the results of these analyses. Although thedescriptive analyses indicated that there was a difference inhospital LOS between the VAP and the non-VAP groups,these differences did not reach statistical significance whenusing Mann–Whitney U analyses, z(38) = −0.069, p = .945.The mean hospital LOS for the VAP group was 14.33 days(±5.701), whereas hospital LOS for the non-VAP group was15.45 days (±10.322).

Analyses were also performed to identify if there weredifferences for PICU LOS between the VAP and the non-VAP group. The mean PICU LOS for the VAP group was11.22 days (±4.893), whereas PICU LOS for the non-VAPgroup was 11.31 days (±8.937). Mann–Whitney U analysesindicated that these differences were not statisticallysignificant, z(38) = −0.844, p = .398.

Results of Regression Analyses: Risk Factors

Several of the study variables (sedation and paralytic use,GI feeds, GCS and Riker scores, oral care, frequency of

Table 3 Differences in Hospital and PICU LOS BetweenGroups

⁎Results not statistically significant. For hospital LOS, p = .945; forPICU LOS, p = .398.

Table 4 Results of Logistic Regression Model

Variable t p

Continuous sedation 0.33 .5638Intermittent sedation 0.31 .5752Paralytics 0.69 .4054GI feeds 0.53 .4679GCS score 0.47 .4946RIKER 1.24 .2657Oral care 0.60 .4383Patient repositioning 0.05 .8248GI decompression 0.03 .8546Use of SCDs 1.14 .2860HOB elevated 0.73 .3930Suctioned every 2 hours 2.26 .1327Cuff checks every 2 hours 0.56 .7576

repositioning, GI decompression, SCDs, HOB elevation,suctioning frequency, and frequency of cuff checks) weregathered at multiple time points (i.e., daily during the PICUstay for each patient). Data on these variables were exploredusing repeated-measures logistic regression modeling tech-niques. Table 4 shows the results from these analyses. Noneof the variables explored were found to be statisticallysignificant risk factors for VAP.

VAP Prevention Interventions

The secondary aim of this study was to identify nursingand respiratory interventions routinely performed toprevent VAP. A list of these interventions was compiledbased on an extensive literature review as to interventionsshown to prevent VAP in the adult population (Abdel-Gaward, El-Hodhod, Ibrahim, & Michael, 2009; Alumu-neef et al., 2004; ATS & IDSA, 2005; Bigham et al.,2009; Bonten et al., 2004; Collard et al., 2003; Curley etal., 2006; daSilva et al., 2010; Fulbrook & Mooney, 2003;Hsieh et al., 2010; Patra et al., 2007; Taira et al., 2009;Yuan et al., 2007). Nursing interventions included oralcare every 2 hours, patient repositioning every 2 hours,HOB elevation, suctioning method (open vs. closed), andsuctioning every 2 hours (ATS & IDSA, 2005; Curleyet al., 2006). Respiratory interventions were cuff pressurechecks every 2 hours, ventilator circuit changes, andventilator condensation removal (ATS & IDSA, 2005;Bonten et al., 2004; Collard et al., 2003). As part ofthis study, we aimed to identify which of these interven-tions recommended in adult VAP prevention werealready being performed and documented in ventilatedpediatric patients. Table 5 presents the results of thesedescriptive analyses.

As shown in Table 5, many of these interventions werepoorly documented. Documentation of most interventionsaccording to VAP prevention guidelines for adults rangedfrom 4% to 47%. Interventions documented specifically by

Table 5 Documentation of VAP Prevention Interventions

6 J. Morinec et al.

nursing staff included oral care, repositioning, HOBelevation, suctioning method, and suctioning frequency. Ofthese interventions, patient repositioning every 2 hours wasrecorded most often (37.2%), followed by oral care every 2hours (22.6%). Suctioning frequency and method and HOBelevation were only recorded 4%–12% of the time.

Interventions routinely performed and documentedby respiratory staff included ETT cuff pressure checks,ventilator circuit changes, and removal of condensation fromthe ventilator tubing. Of these interventions, cuff pressurechecks were documented most often (46.2%). The remainingtwo interventions were rarely recorded (2.4%).

Collectively, these nursing and respiratory interventionsare important in preventing VAP. It is believed that most ofthese interventions are routinely performed by nursing andrespiratory staff when caring for ventilated pediatric patients.However, improvement in the actual documentation of theseinterventions is needed.

Summary

This study aimed to identify the risk factors associatedwith VAP among ventilated pediatric patients and to providebaseline data on current VAP prevention practices in a PICU.Although the risk factors included in this study did not proveto be statistically significant in predicting VAP, findings docontribute information about the current practices for VAP

prevention in ventilated pediatric patients. This is particu-larly important as clinicians make decisions regardingimplementation of all or part of VAP bundles that arecurrently used in the adult population.

In this study, data on the frequency of nursing andrespiratory interventions to prevent VAP provide informa-tion on current practices. First, data suggest that ventilatedpediatric patients are receiving adequate sedation, asevidenced by the high percentage of patients with continuousand/or intermittent sedation. In the adult population, a keycomponent of the VAP bundle is daily sedation interruptions(i.e., “sedation vacations”) to aid in identifying readiness forextubation, which may decrease ventilator days, LOS, andthe risk for VAP (Curley et al., 2006). Currently, there are nodata to support recommendation of sedation vacations inpediatrics, and it is unclear if this intervention should beincorporated as part of a VAP bundle in this population.There was no documentation of daily sedation interruptionsin our study, and this would be an important area foradditional research.

Second, results from our study indicate that the use ofneuromuscular blockade is infrequent, as evidenced by thelow rate of paralytic use within the study sample. In a studyby daSilva et al. (2010), data were gathered on 317 pediatricpatients, and those with sustained neuromuscular blockadehad longer duration of mechanical ventilation and PICULOS and a higher incidence of VAP than did controls. Theneuromuscular blockade group was composed of only 34patients, which was approximately 10% of the study sample.

7VAP Risk Factors and Interventions

Thus, although it was concluded that neuromuscularblockade was a risk factor for VAP, it was not commonlyused. Our study was similar in the rare use of neuromuscularblockade; however, we did not find this to be a risk factorfor VAP.

Third, our study found that 93% of patients had NGTs inplace.Of this number, just less than half (45%) received enteralfeedings, whereas the remaining participants (53%) had theNGTs to suction to provide gastric decompression. Mostpatients (69%) received some type of GI ulcer prophylaxis.This finding is consistent with the mixed literature in both theadult and pediatric literature on the effectiveness of GI feeding,decompression, and ulcer prophylaxis on the development ofVAP (Abdel-Gaward et al., 2009; Alumuneef et al., 2004;Reviez, Guerrero-Lozano, Camacho, Yara, & Mosquera,2010; Yildizdas et al., 2002). A recent systematic review ofstress ulcer, gastritis, and GI bleeding in critically illpediatric patients indicates that there may be no significantdifferences in the development of pneumonia when patientsare or are not given GI prophylaxis (Reviez et al., 2010),yet GI reflux is common among ventilated pediatricpatients (Abdel-Gaward et al., 2009). Although theliterature may not yield sufficient evidence in support ofthese interventions, findings from our study indicate thatthey may be commonly done. Additional research is neededto better determine if GI feeding, decompression, or ulcerprevention should be necessary components in futurepediatric VAP bundles.

Another key finding from our study is that documen-tation of nursing and respiratory interventions to preventVAP is crucial. Bigham et al. (2009) identified this as animportant factor in tracking compliance with VAPbundles. In fact, multiple measures including checklistsand random audits were put into place in the Bighamproject to ensure adequate documentation of VAP bundleinterventions. Our study findings indicate that many ofthese interventions are not routinely documented. ETTcuff pressure checks and patient repositioning wereamong the most frequently documented interventions,yet documentation rates of these interventions were onlyapproximately 40%, which is not adequate whendetermining compliance with practice recommendations.The remaining nursing and respiratory interventions wererarely documented, regardless if they were believed bythe research staff to be performed or not by clinicians.Documentation of these VAP prevention efforts will needto be a key component in any future VAP bundle or VAPprevention efforts in pediatrics to track compliance andmake adequate inferences about the effectiveness of theseinterventions on VAP rates.

In summary, research is just beginning to identify thefactors associated with VAP in ventilated pediatric patients.There is a lack of clear evidence-based guidelines to guidePICU practice in this area. Research has provided data onpossible risk factors for VAP in the pediatric setting, andmany of these risk factors are similar to those in adults.

However, additional research is needed to substantiate theseearly findings and to determine the appropriate componentsof a PICU VAP bundle.

Our study was limited by a small sample size and theuse of a single site for data collection. The small samplesize may have contributed to the possibility of a Type IIerror, which could explain why no risk factors werefound to be statistically significant in our study. Findingsdo, however, provide information about current practicesto prevent VAP in the pediatric setting and highlight theimportance of documentation to demonstrate compliancewith VAP prevention efforts. Documentation of theseefforts will be a key component in future VAPprevention practices.

Acknowledgments

This study did not receive any extramural funding. Thefindings from this study have been accepted for a posterpresentation at the Society for Pediatric Nursing's annualconference, April 29–May 2, 2010, Orlando, FL. There is nocommercial financial support associated with this study.

References

Abdel-Gaward, T. A., El-Hodhod, M. A., Ibrahim, H. M., & Michael, Y. W.(2009). Gastroesophageal reflux in mechanically ventilated pediatricpatients and its relation to ventilator associated pneumonia. CriticalCare Medicine, 13, R164.

Alumuneef, M., Memish, Z. A., Balkhy, H. H., Alalem, H., & Abutaleb, A.(2004). Ventilator-associated pneumonia in a pediatric intensive careunit in Saudi Arabia: A 30-month prospective surveillance. InfectiousControl and Hospital Epidemiology, 25(9), 753−758.

American Thoracic Society (ATS), & The Infectious Disease Society ofAmerica (IDSA). (2005). Guidelines for the management of adults withhospital-acquired, ventilator-associated, and healthcare-associated pneu-monia. American Journal of Respiratory and Critical Care Medicine,171, 388−416.

Bigham, M. T., Amato, R., Bondurrant, P., Fridriksson, J., Krawczewski,C. D., Raake, K., et al. (2009). Ventilator-associated pneumonia in thepediatric intensive care unit: Characterizing the problem and imple-menting a sustainable solution. Journal of Pediatrics, 154, 582−587.

Bonten, M., Kollef, M., & Hall, J. (2004). Risk factors for ventilator-associated pneumonia: From epidemiology to patient management.Clinical Infectious Disease, 38, 1141−1149.

Collard, H., Saint, S., & Matthay, M. (2003). Prevention of ventilator-associated pneumonia: An evidence-based systematic review. Annals ofInternal Medicine, 138, 494−501.

Curley, M., Schwalenstocker, E., Deshpande, J., Ganser, C., Bertoch, D.,Brandon, J., et al. (2006). Tailoring the Institute for HealthcareImprovement 100,000 Lives Campaign to pediatric settings: Theexample of ventilator-associated pneumonia. Pediatric Clinics ofNorth America, 53, 1231−1251.

daSilva, P. S., Neto, H. M., deAguiar, V. E., Lopes, E., & deCarvalho, W. B.(2010). Impact of sustained neuromuscular blockade on outcome ofmechanically ventilated children. Pediatric International Retrievedfrom http://www.ncbi.nlm.nih.gov/pumed/20202154.

Evans, B. (2005). Best-practice protocols: VAP prevention. NursingManagement, 36, 10−16.

8 J. Morinec et al.

Foglia, E., Meier, M. D., & Elward, A. (2007). Ventilator associatedpneumonias in neonatal and pediatric intensive care unit patients. Cli-nical Microbiology Reviews, 20, 409−425.

Fulbrook, P., &Mooney, S. (2003). Care bundles in critical care: A practicalapproach to evidence-based practice. British Association of CriticalCare Nurses Nursing in Critical Care, 8, 249−255.

Hsieh, T., Wu, C., Lin, T., Chang, C., & Wong, K. (2010). Frequency ofventilator-associated pneumonia with 3-day versus 7-day ventilatorcircuit changes. Pediatrics and Neonatology, 51, 37−43.

Lachman, P., & Yuen, S. (2009). Using care bundles to prevent infection inneonatal and paediatric ICUs. Current Opinion in Infectious Diseases,22, 224−228.

Patra, P. K., Jayashree, M., Singhi, S., Ray, P., & Saxena, A. K. (2007).Nosocomial pneumonia in a pediatric intensive care unit. IndianPediatrics, 44, 511−518.

Powers, J. (2006). Managing VAP effectively to optimize outcomes andcosts. Nursing Management, 37, 48−54.

Resnar, R., Pronovost, P., Haraden, C., Simmonds, T., Rainey, T., & Nolan,T. (2005). Using a bundle approach to improve ventilator care processesand reduce ventilator associated pneumonia. Joint Commission Journalof Quality and Patient Safety, 31, 243−248.

Reviez, L., Guerrero-Lozano, R., Camacho, A., Yara, L., & Mosquera, P.(2010). Stress ulcer, gastritis, and gastrointestinal bleeding prophylaxis

in critically ill pediatric patients: A systematic review. Pediatric CriticalCare Medicine, 11, 124−132.

Tablan, O., Anderson, L., Besser, R., Bridges, C., & Hajjeh, R.(2003). Guidelines for preventing health care associated pneumonia.2003. Retrieved from http://cdc.gove/mmwr/preview/mmwrhtml/rr5303al.htm.

Taira, B. R., Fenton, K. E., Lee, T. K., Meng, H., McCormack, J. E., Huang,E., et al. (2009). Ventilator-associated pneumonia in pediatric traumapatients. Pediatric Critical Care Medicine, 10, 491−494.

Turton, P. (2008). Ventilator-associated pneumonia in paediatric intensivecare: A literature review. Nursing Critical Care, 13, 214−218.

Yildizdas, D., Yapicioglu, H., & Yilmaz, H. (2002). Occurrence ofventilator-associated pneumonia in mechanically ventilated pediatricintensive care patients during stress ulcer prophylaxis withsucralfate, ranitidine, and omeprazole. Journal of Critical Care,17, 240−245.

Youngquist, P., Carroll, M., Farber, M., Macy, D., Madrid, P., Ronning, J.,et al. (2007). Implementing a ventilator bundle in a communityhospital. Joint Commission Journal of Quality and Patient Safety, 33,219−225.

Yuan, T. M., Chen, L. H., & Yu, H. M. (2007). Risk factors and outcomesfor ventilator-associated pneumonia in neonatal intensive care patients.Journal of Perinatology Medicine, 35, 334−338.