An epidemiological study of ventilator-associated ...
Transcript of An epidemiological study of ventilator-associated ...
An epidemiological study of ventilator-associated pneumonia (VAP) and
ventilator-associated events (VAE) surveillance and preventative strategies
in critically ill children
Noor Azizah Mohd Ali
RN BHSc (Nursing) (Hons) MNSc Cert (Teaching)
A thesis submitted for the degree of Doctor of Philosophy at
The University of Queensland in 2019
School of Nursing, Midwifery, and Social Work
i
Abstract
Background: Ventilator-associated pneumonia (VAP) is an ongoing iatrogenic burden within the
healthcare system for both adults and paediatrics. Debate continues over the appropriateness of the
VAP surveillance tool in paediatrics. Implementation of the new ventilator-associated events (VAE)
surveillance tool in the adult setting have shown to overcome the subjectivity of the traditional
pneumonia 1 VAP (PNU1/VAP) surveillance tool. Unlike in adult units, the application of the VAE
surveillance tool in paediatrics has not been mandated, leaving a question as to its potential
application. The lack of VAP paediatric specific studies has hampered progress informing compliance
with VAP preventative strategies in paediatric intensive care units (PICUs). Despite good hand
hygiene practise being established as a vital infection control measure, perception of ‘Speaking up
for hand hygiene’ in PICU still under- research. Hence, it became a potential area to improve good
hygiene practice.
Aim: To determine the VAP and VAE incidence using two surveillance tools at two time points; at
baseline and post an education campaign aimed at both PICU staff and parents. In addition to testing
the two surveillance tools, compliance auditing and staff and parental perspectives of VAP/VAE and
preventative strategies was established.
Methods: Retrospective study: PNU1/VAP and VAE surveillance tools were applied to 262
mechanical episodes of 234 children who received invasive mechanical ventilation ≥48 hours in PICU
in 2015. The sensitivity and specificity of VAE surveillance tool was tested. Other epidemiological
data were recorded; demographic characteristics, risk factors and VAP preventative strategies
documented within the unit.
VAP education, VAP compliance auditing with feedback and surveys: reinforcement of updated VAP
education was launched for PICU staff, and a pamphlet focusing on ‘Speaking up for hand hygiene’
was developed primarily to educate parents. The information was delivered to parents (N=37) via a
pamphlet and face-to-face education. VAP compliance auditing was conducted for a two-month
period and involved 37 patients in PICU undergoing VAP preventative strategies implemented by
PICU staff and parents: hand hygiene, oral hygiene, endotracheal tube (ETT) suctioning, ETT cuff
pressure checks, head of bed elevation, ventilator circuit checks, and early enteral feeding
commencement. The parents’ and nurses’ perceptions of ‘Speaking up for hand hygiene’ were
examined through surveys undertaken by 19 parents and 34 nurses.
ii
Prospective study: a six-month prospective study was conducted to investigate any improvement in
incidence rates of VAP/VAE and estimate the compliance rate of VAP preventative strategies.
Results: The incidence rate of VAP/VAE was 9.3 per 1000 ventilator days (VAP/VAE) based on end
of ventilation and 10.2 (VAP) and 10.4 (VAE) per 1000 ventilator days based on when the patients
were no longer at risk. The specificity of the new VAE surveillance tool was high with a slight
agreement between the tools. The overall compliance with VAP preventative strategies was 89.0%.
The presence of gastrointestinal prophylaxis (GI) and the frequency of oral hygiene were predictors
of the potential incidence rate of VAE/hour of ventilation, but none were found for VAP.
Overall VAP preventative strategies compliance was measured at 83.1%. Hand hygiene compliance
among PICU staff was at >80% and 64.7% among parents. Oral hygiene and ETT cuff pressure
checks (sub-elements) reported a compliance rate of <80.0%. Parents and nurses agreed that the
‘Speaking up for hand hygiene’ initiative would increase hand hygiene practises and willingness to
be reminded to perform the practise in PICU. Some parents reported reason being at vulnerable
position to questioning, made them hesitate to remind nurses and other PICU staff to perform hand
hygiene. Nurses reported their concern for the parents’ emotional status and preconceptions that their
colleagues were unwilling to accept hand hygiene reminders as reasons for not reminding.
The reduction of the incidence rate for VAP/VAE showed a drop to 3.9 (VAP) and 4.4 (VAE) and to
2.8 (VAP) and 3.2 (VAE) per 1000 ventilator days based on end of ventilation and until the patients
were no longer at risk. There was a statistically significant improvement of VAP preventative
strategies’ compliance reported between prospective and retrospective studies.
Conclusions: The reduction of VAP/VAE incidence rates demonstrated in this study reinforces the
need for VAP education, compliance auditing with feedback, and hand hygiene education for parents
and staff. Surveys on ‘Speaking up for hand hygiene’ among parents and nurses provided improved
perceptions which may help promote infection control measures in PICUs. Although the VAE tool
had slight agreement with the PNU1/VAP tool, it suggested the merit of identification of non-VAP
complications through a high specificity result. No risk factors or VAP preventative strategies were
found to be predictive of VAP, although these factors may be potential factors for non-infectious
complications. The presence of GI prophylaxis and frequency of oral hygiene performance were
found to be associated with VAE development. These findings were not robust, due to the low events
rate, but they were worthwhile predictors to be researched further.
iii
Declaration by author
This thesis is composed of my original work, and contains no material previously published or written
by another person except where due reference has been made in the text. I have clearly stated the
contribution by others to jointly-authored works that I have included in my thesis.
I have clearly stated the contribution of others to my thesis as a whole, including statistical assistance,
survey design, data analysis, significant technical procedures, professional editorial advice, financial
support and any other original research work used or reported in my thesis. The content of my thesis
is the result of work I have carried out since the commencement of my higher degree by research
candidature and does not include a substantial part of work that has been submitted to qualify for the
award of any other degree or diploma in any university or other tertiary institution. I have clearly
stated which parts of my thesis, if any, have been submitted to qualify for another award.
I acknowledge that an electronic copy of my thesis must be lodged with the University Library and,
subject to the policy and procedures of The University of Queensland, the thesis be made available
for research and study in accordance with the Copyright Act 1968 unless a period of embargo has
been approved by the Dean of the Graduate School.
I acknowledge that copyright of all material contained in my thesis resides with the copyright
holder(s) of that material. Where appropriate I have obtained copyright permission from the copyright
holder to reproduce material in this thesis and have sought permission from co-authors for any jointly
authored works included in the thesis.
iv
Publication during candidature
Peer-reviewed paper
Mohd Ali, N. A., Jauncey-Cooke, J., & Bogossian, F. (2019). Ventilator-associated events in children:
A review of literature. Australian Critical Care, 32(1), 55-62. doi:10.1016/j.aucc.2018.11.063
Conference abstract
Mohd Ali, N. A., Bogossian, F., Jauncey-Cooke, J., & Ballard, E. (2017). Preventative strategies of
VAP: Lessons from a one-year retrospective study. Poster presented at the 42nd Australia and
New Zealand Annual Scientific Meeting on Intensive Care and the 23rd Annual Paediatric and
Neonatal Intensive Care Conference. Gold Coast, Australia, 11 October to 13 October.
v
Publications included in this thesis
Mohd Ali, N. A., Jauncey-Cooke, J., & Bogossian, F. (2019). Ventilator-associated events in children:
A review of literature. Australian Critical Care, 32(1), 55-62. doi:10.1016/j.aucc.2018.11.063-
incorporated in Chapter 2.
Contributor Statement of contribution
Author Noor Azizah Mohd Ali Critically reviewed the paper (80%)
Wrote the paper (70%)
Author Jaqueline Jauncey-Cooke Critically reviewed the paper (20%)
Wrote and edited the paper (20%)
Author Fiona Bogossian Wrote and edited the paper (10%)
vi
Contributions by others to the thesis
Dr Jacqueline Jauncey-Cooke (Principal Advisor), Honorary Prof Fiona Bogossian (Associate
Advisor) and Dr Emma Ballard (Associate Advisor) provided substantial guidance and inputs into
development of this thesis and throughout the PhD candidature. All of the supervisors critically
reviewed and provided comprehensive feedback on the content of this thesis.
Statement of parts of the thesis submitted to qualify for the award of another degree
None
vii
Acknowledgements
By the name of Allah, the most compassionate and the most merciful, and blessing to Prophet
Muhamad s.a.w. All praises to Him who has given me strength to embark on this PhD journey. May
I be able to serve back society through the knowledge that I have gained during this journey. My
heartfelt appreciation goes to my advisory team, Dr. Jacqueline Jauncey-Cooke (principal advisor),
Honorary Prof. Fiona Bogossian (associate advisor), and Dr. Emma Ballard (associate advisor) for
their patience, guidance, encouragement, and meaningful suggestions from the beginning until the
end of this journey. I gratefully acknowledge the constructive comments and suggestions from
reviewers throughout my milestones (Dr. Sara Mayfield, Dr. Karen New, and Dr. Haakan Strand).
I also would like to express sincere thanks to A/Prof. Dr. Luregn Schlapbach, Ivy Chang, Andrew
Barlow, Kylie Pearson, Dr. Deborah Long, Liz Crowe, Marina Demosthenous, and Tara Williams for
their enormous support, especially during the preliminary stage of my proposal and throughout the
data collection process in the Queensland Children’s Hospital. I would like to give a heartfelt thanks
to all the PICU staff for their support, and a special thanks to the staff of the Centre for Children’s
Health Research Centre (CHRC) who provided me a work station close to the hospital during my data
collection.
Thank you to my family, my father Mohd Ali binYusof, and my mother Siti Selahah binti Ali for
their prayers, support, advice, and motivation. I would especially like to thank the Malaysian
Government and my employer, the International Islamic University of Malaysia, for their financial
support and study leave approval. I would further like to thank my fellow colleagues in the School of
Nursing, Midwifery and Social Work and fellow friends (Dr Alison Williams, Dr Adrienne Hudson
and families) for their support through the good times and the bad. Thank you to all who supported
me in writing and encouraged me to strive towards my goal.
Terima kasih.
viii
Keywords
children, paediatric, epidemiology, ventilator associated pneumonia, ventilator associated events,
surveillance, preventative strategies
Australian and New Zealand Standard Research Classifications (ANZSRC)
ANZSRC code: 111002, Clinical Nursing: Primary (Preventative), 25%
ANZSRC code: 111403 Paediatrics, 25%
ANZSRC code: 110003, Clinical Nursing: Secondary (Acute Care), 20%
ANZSRC code: 111706, Epidemiology, 30%
Fields of Research (FoR) Classification
FoR code: 1110, Nursing, 50%
FoR code: 1117, Public Health and Health Services, 50%
ix
This thesis dedicated to my niece, Nurbatrisyia Khairina Khairul Anuar
x
Table of Contents
Abstract ………………………………………………………………………………………….i
Acknowledgements........................................................................................................................... vii
Table of Contents ............................................................................................................................... x
List of Figures ................................................................................................................................. xvii
List of Tables .................................................................................................................................xviii
List of Abbreviation ......................................................................................................................... xx
Thesis Introduction ...................................................................................................... 1
Introduction .......................................................................................................................... 1
Mechanical ventilation and VAP and VAE ......................................................................... 1
Aim ....................................................................................................................................... 5
Research questions ............................................................................................................... 5
Significance of the study ...................................................................................................... 6
Definitions ............................................................................................................................ 7
The flow of the thesis ........................................................................................................... 8
Summary ............................................................................................................................ 10
Literature Review: Epidemiology of VAP and VAE in paediatrics ...................... 11
Introduction to the literature review chapters .................................................................... 11
Timeline of surveillance of pneumonia (PNU) in healthcare settings ............................... 11
VAP and VAE: Pathogenesis and factors for acquiring VAP/VAE .................................. 14
The global incidence and the impact of VAP and VAE .................................................... 15
Risk factors for VAP and VAE .......................................................................................... 17
Diagnostic criteria for pneumonia (PNU/VAP) ................................................................. 20
The CDC alternative criteria for infants and children (PNU1) ...................................... 20
Diagnostic criteria for VAE surveillance ........................................................................... 21
Ventilator associated events (VAE) in children: A review of literature ............................ 24
Publication ..................................................................................................................... 24
Summary ............................................................................................................................ 35
Literature Review: Preventative strategies, compliance and VAP education ...... 37
Introduction ........................................................................................................................ 37
xi
Preventative strategies for VAP ......................................................................................... 37
Evidence of individual preventative strategies within the VAP bundle ............................ 41
Hand Hygiene ................................................................................................................ 41
Oral Hygiene .................................................................................................................. 42
Endotracheal Suctioning ................................................................................................ 44
Head of the bed (HOB) Elevation .................................................................................. 46
Endotracheal tube (ETT) and cuff pressure checks ....................................................... 46
Ventilator circuit checks ................................................................................................ 47
Sedation interruptions .................................................................................................... 48
Early enteral feeding commencement ............................................................................ 49
Gastrointestinal (GI) prophylaxis................................................................................... 50
Summary ........................................................................................................................ 51
Organisational and clinician compliance with VAP preventative strategies ..................... 51
Improving compliance with VAP preventative strategies via VAP education .................. 53
VAP Education .............................................................................................................. 53
The role of parents in VAP prevention and parental education on hand hygiene.............. 57
Perceptions of ‘Speaking up for hand hygiene’ among healthcare workers and parents ... 58
VAE and its preventative strategies ................................................................................... 61
Summary ............................................................................................................................ 61
Research Methodology .............................................................................................. 62
Introduction ........................................................................................................................ 62
Study setting ....................................................................................................................... 62
Study design ....................................................................................................................... 62
Phase 1: Retrospective study ......................................................................................... 63
Population and sample .......................................................................................... 64
Data collection ...................................................................................................... 64
VAP and VAE surveillance tools ......................................................................... 65
Demographic characteristics and admission-related variables ............................. 65
Potential risk factors ............................................................................................. 65
VAP preventative strategies ................................................................................. 65
Data analysis ......................................................................................................... 66
xii
Assessment of compliance of VAP preventative strategies ................................. 71
Ethical considerations ........................................................................................... 72
Phase 2: VAP education, VAP preventative strategies compliance auditing and surveys
………………………………………………………………………………………..72
VAP education and education engagement .......................................................... 73
Compliance auditing ............................................................................................. 80
Survey for parents ................................................................................................. 85
Surveys for nurses................................................................................................. 89
Phase 3: Prospective study ............................................................................................. 92
Population and sample .......................................................................................... 92
Data collection ...................................................................................................... 92
VAP and VAE surveillance tools ......................................................................... 93
Data analysis ......................................................................................................... 93
Ethical considerations ........................................................................................... 93
4.4 Summary .................................................................................................................................. 94
Results and discussion Phase 1: Retrospective study ............................................. 95
Introduction ........................................................................................................................ 95
Selection of patients with eligible mechanical ventilation episodes .................................. 95
Demographic characteristics of patients in PICU admission ......................................... 96
Outcome variables according to mechanical ventilation episodes ................................ 97
VAP and VAE counts according to mechanical ventilation episodes ........................... 98
Incidence and prevalence of VAP in PICU ................................................................... 98
Incidence and prevalence of VAE in PICU ................................................................... 99
The sensitivity and the specificity of the VAE surveillance tool ................................... 99
Agreement between the two surveillance tools............................................................ 100
Discussion .................................................................................................................... 100
Discussion of demographic characteristics of the patients ................................. 100
Discussion of VAP and VAE in the PICU of the QCH in 2015......................... 101
Potential risk factors for VAP and VAE .......................................................................... 103
Compliance of preventative strategies for VAP............................................................... 103
xiii
Univariate analysis ........................................................................................................... 104
Association between demographic characteristics and VAP and VAE ....................... 104
Association of outcome characteristics with VAP and VAE ....................................... 105
Association of possible risk factors with VAP and VAE ............................................ 106
Association of preventative strategies with VAP and VAE ........................................ 107
Summary of the significant explanatory variables found in the study ......................... 108
Discussion .................................................................................................................... 109
Possible risk factors and compliance of VAP preventative strategies ................ 109
Discussion of the association between study variables and VAP/VAE development
at univariate analysis level ................................................................................................... 110
Multivariate analysis ........................................................................................................ 112
VAP models ................................................................................................................. 112
Age and gender adjusted Poisson and negative binominal models .................... 112
VAE models ................................................................................................................. 112
Age and gender adjusted Poisson and negative binominal models .................... 112
The final modified Poisson and Negative Binominal models for VAE ............. 116
Discussion .................................................................................................................... 119
Discussion of potential risk factors and VAP preventative strategies and
association with VAP/VAE development by multivariate analysis ..................................... 119
Summary .......................................................................................................................... 120
Results and discussion Phase 2: VAP preventative strategy compliance and
surveys ................................................................................................................................. 122
Introduction ...................................................................................................................... 122
Results of compliance auditing of VAP preventive strategies ......................................... 122
Demographic characteristics of participants in VAP preventative strategy compliance
auditing ........................................................................................................................ 122
Hand hygiene ...................................................................................................... 122
Oral hygiene........................................................................................................ 123
Cuff pressure check ............................................................................................ 123
Endotracheal suctioning ...................................................................................... 123
Head of bed (HOB) elevation ............................................................................. 124
xiv
Ventilator circuits checks ................................................................................... 124
Enteral feeding commencement within 24 hours of admission .......................... 124
Discussion of compliance of VAP preventative strategies during auditing................. 124
Parental survey: ‘Speaking up for hand hygiene’ ........................................................ 127
Response rate ...................................................................................................... 127
Demographic characteristics of parents and their child’s admission history ..... 128
Perceptions of parents on VAP education .......................................................... 128
Parental perceptions on willingness to remind and to be reminded by nurses and
other PICU staff to perform hand hygiene ........................................................................... 128
Reasons parents would be reluctant to prompt nursing staff regarding hand hygiene
………………………………………………………………………………….128
Reasons parents would be reluctant to remind other PICU staff regarding hand
hygiene 129
Suggestions or comments to improve hand hygiene practise in the unit ............ 129
Discussion of parental perceptions on ‘Speaking up for hand hygiene’............. 129
Nursing staff survey ..................................................................................................... 132
Response rate ...................................................................................................... 132
Demographic characteristics of nursing staff participants .................................. 132
Nurses’ perceptions of their willingness to remind and be reminded by parents and
other PICU staff to perform hand hygiene ........................................................................... 132
Reasons nurses would be reluctant to prompt parents regarding hand hygiene . 132
Reasons nurses would be reluctant to prompt other PICU staff regarding hand
hygiene ……… .................................................................................................................... 133
Suggestions or comments to improve hand hygiene practise in the unit ............ 133
Discussion of nurses’ perceptions on ‘Speaking up for hand hygiene’ .............. 133
Summary .......................................................................................................................... 135
Results and discussion Phase 3: Prospective Study .............................................. 136
Introduction ...................................................................................................................... 136
Results .............................................................................................................................. 136
Baseline demographic characteristics .............................................................................. 136
xv
Demographic characteristics of patient based on PICU admission ............................. 136
Outcome variables according to mechanical ventilation episodes at prospective study....
………………………………………………………………………………………..138
Incidence of VAP and VAE at prospective study ........................................................ 139
Discussion .................................................................................................................... 140
Discussion of demographic and outcome variables of the prospective study .... 140
Discussion of incidence of VAP and VAE at prospective study ........................ 140
Potential risk factors according to mechanical ventilation episodes............................ 142
Compliance of preventative strategies of VAP in the prospective study ..................... 143
Compliance of VAP preventative strategies: Retrospective versus prospective
studies …….......................................................................................................................... 147
Discussion of VAP preventative strategies compliance — retrospective versus
prospective studies ....................................................................................................... 147
Univariate analysis ........................................................................................................... 148
Association of demographic characteristics with VAP and VAE in the prospective
study ............................................................................................................................. 149
Association of patient outcome characteristics with VAP and VAE ........................... 150
Association of possible risk factors with VAP and VAE ............................................ 150
Association of preventative strategies with VAP and VAE ........................................ 151
Discussion of demographic and outcome characteristics and VAP/VAE — prospective
versus retrospective studies .......................................................................................... 152
Discussion of potential risk factors and preventative strategies and VAP/VAE —
prospective versus retrospective studies ...................................................................... 153
Summary .......................................................................................................................... 154
General Discussion ................................................................................................... 155
Introduction ...................................................................................................................... 155
Summary of the thesis ...................................................................................................... 155
New VAE surveillance tool for global surveillance ........................................................ 158
The role of VAP education, compliance auditing with feedback, and parents’ involvement
in VAP/VAE prevention .................................................................................................. 160
Limitations and strengths ................................................................................................. 161
General implications and future directions ...................................................................... 164
xvi
Conclusion........................................................................................................................ 165
List of References ........................................................................................................................... 167
Appendices 188
Appendix A Data collection sheet (Retrospective study) ........................................................... 188
Appendix B HREC approval ........................................................................................................ 200
Appendix C Governance approval ............................................................................................... 203
Appendix D PHA approval ........................................................................................................... 204
Appendix E University of Queensland Ethical approval ........................................................... 205
Appendix F Updated VAP education package (PICU staff) ..................................................... 206
Appendix G Education for parents: Pamphlet ........................................................................... 207
Appendix H Survey Questionnaire (Parents) .............................................................................. 209
Appendix I Survey participant information sheets (Parents and Nurses) .............................. 213
Appendix J Survey Questionnaire (Nurses) ............................................................................... 217
Appendix K Data collection sheet (Prospective study) ............................................................... 219
Appendix L Parents information sheet and consent form for prospective study .................... 234
Appendix M Approval of waiver of consent for prospective study ........................................... 237
xvii
List of Figures
Figure 1.1: Flow of thesis presentation ................................................................................................ 9
Figure 2:1: Timeline for surveillance of pneumonia (PNU) in a healthcare setting .......................... 13
Figure 2:2: Ventilator-associated events in the respective tiers ......................................................... 15
Figure 2:3: PRISMA flow diagram of literature selection ................................................................. 27
Figure 4:1: Study timeframe with respective research activities undertaken .................................... 62
Figure 4:2: The research components involved in Phase 2 of the study ............................................ 73
Figure 5:1: 262 episodes of mechanical ventilation met the study criteria ........................................ 96
Figure 6:1: Endotracheal suctioning (open method) compliance .................................................... 123
Figure 7:1: Comparison of individual VAP preventative strategy compliance — retrospective versus
prospective studies; p-values given where change is statistically significant. .............. 147
xviii
List of Tables
Table 2.1: Risk factors for VAP in children: Studies published from 2005 to 2015 ......................... 19
Table 2.2: CDC PNU1/VAP Tool: Alternative Criteria for Infants and Children ............................. 22
Table 2.3: The CDC VAE tiers with respective requirements and criteria ........................................ 23
Table 2.4: Study summaries and agreement between surveillance tools ........................................... 28
Table 3.1: Additional preventative strategies to consider for the paediatric VAP bundle ................. 38
Table 3.2: Individual VAP preventative strategies used in paediatrics and VAP rates (in order of
publication date) .............................................................................................................. 39
Table 3.3: Oral hygiene protocol for mechanically ventilated children ............................................ 43
Table 3.4: Paediatric studies involving VAP education and VAP preventative strategy
implementation ................................................................................................................ 55
Table 4.1: Compliance standard for VAP preventative strategies according to PICU and national
standard for hand hygiene ............................................................................................... 71
Table 4.2: The content of updated VAP education in PICU .............................................................. 74
Table 4.3: The content of VAP preventative strategies poster .......................................................... 75
Table 4.4: The summary of content validation for bi-fold pamphlet, ‘VAP: How I Can Help my Child
in PICU’ .......................................................................................................................... 77
Table 4.5: The VAP education engagement in PICU at Phase 2 ....................................................... 79
Table 4.6: VAP compliance data auditing ......................................................................................... 82
Table 4.7: Elements in parents’ survey .............................................................................................. 88
Table 4.8: Elements in nurses’ survey ............................................................................................... 91
Table 5.1: Demographic characteristics according to patient admission (n=253) ............................. 97
Table 5.2: Outcome variables according to mechanical ventilation episodes (n=262) ..................... 98
Table 5.3: Contingency table of VAE versus VAP ........................................................................... 99
Table 5.4: Risk factors according to mechanical ventilation episodes (n=262) .............................. 103
Table 5.5: Comparison between VAP preventative strategies compliance and PICU/national standard
practise .......................................................................................................................... 104
Table 5.6: Univariate analysis for association of demographic characteristics with VAP/VAE (n=262)
....................................................................................................................................... 105
Table 5.7: Univariate analysis for association of patient outcome characteristics with VAP and VAE
(n=262) .......................................................................................................................... 106
Table 5.8: Univariate analysis for association of possible risk factors with VAP and VAE (n=262)
....................................................................................................................................... 107
xix
Table 5.9: Univariate analysis for association of VAP preventative strategies with VAP/VAE (n=262)
....................................................................................................................................... 108
Table 5.10: Explanatory variables with a significant association with VAP or VAE by univariate
analysis .......................................................................................................................... 108
Table 5.11: Age and gender adjusted Poisson and Negative Binominal regression models of risk
factors and preventative strategies for incidence of VAP/hour of ventilation .............. 114
Table 5.12: Age and gender adjusted Poisson and Negative Binominal regression models of risk
factors and preventative strategies for incidence of VAE/hour of ventilation .............. 115
Table 5.13: Final Poisson models of risk factors and preventative strategies for incidence of VAE/
hour of ventilation ......................................................................................................... 117
Table 5.14: Final Negative Binominal regression models of risk factors and preventative strategies
for incidence of VAE/hour of ventilation ..................................................................... 118
Table 7.1: Demographic characteristics comparison of patients in prospective study versus
retrospective study on PICU admission (n=115; n= 253) respectively......................... 137
Table 7.2: Outcome variables according to mechanical ventilation episodes prospective versus
retrospective study (n=120; n= 262) ............................................................................. 138
Table 7.3: Comparison of VAP and VAE incidence between retrospective and prospective studies
measured on end of ventilation and until the patient is no longer at risk ...................... 139
Table 7.4: VAP and VAE counts (tiers) according to mechanical ventilation episode prospective and
retrospective studies ...................................................................................................... 140
Table 7.5: Potential risk factors according to mechanical ventilation episode (n= 120) in prospective
study, compared to retrospective study (n= 262) .......................................................... 143
Table 7.6: Comparison of VAP compliance preventative strategy performance with PICU
standard/National standard ............................................................................................ 145
Table 7.7: Univariate analysis for association of demographic characteristics with VAP and VAE
(n=120) .......................................................................................................................... 149
Table 7.8: Univariate analysis for association of outcome characteristics with VAP and VAE (n=120)
....................................................................................................................................... 150
Table 7.9: Univariate analysis for association of possible risk factors with VAP and VAE (n=120)
....................................................................................................................................... 151
Table 7.10: Univariate analysis for association of VAP preventative strategies with VAP and VAE
(n=120) .......................................................................................................................... 152
xx
List of Abbreviation
CDC Centre for Disease Control and Prevention
CI Confidence Interval
FiO2 Fraction Inspired Oxygen
HCAIs Healthcare-Associated Infections
HR Hazard Ratio
IHI Institute for Healthcare Improvement
IQR Interquartile Range
IVAC Infection-Related Associated Condition
NHSN National Healthcare Safety Network
OR Odd Ratio
PEEP Positive End Expiratory Pressure
PICU Paediatric Intensive Care Unit
PNU Pneumonia
PVAP Possible Ventilator- Associated Pneumonia
SD Standard deviation
VAC Ventilator-Associated Condition
VAE Ventilator-Associated Events
VAP Ventilator-Associated Pneumonia
1
Thesis Introduction
Introduction
This thesis investigates ventilator-associated pneumonia (VAP) and ventilator-associated events
(VAE) in mechanically ventilated children. Within the thesis, the incidence and risk factors of VAP
and VAE are explored using surveillance tools recognised for adults and children. Preventative
strategies for VAP incorporating both staff and parents are implemented and their impact is measured.
This chapter introduces the epidemiology of the healthcare-associated infections (HAIs) of VAP and
VAE and discusses the significance of this study. The research questions, aims, and definitions of
key terms are described.
Mechanical ventilation and VAP and VAE
Mechanical ventilation is a life-saving option when the airway requires protection and respiratory
effort and/or gas exchange is compromised (Arca, Uhing, & Wakeham, 2015). Invasive mechanical
ventilation requires an endotracheal tube as the interface between the ventilator and lungs. The
presence of an endotracheal tube and concomitant sedation impairs the natural mechanisms of
breathing by suppressing reflexes such as coughing and gagging (Mietto, Pinciroli, Patel, & Berra,
2013). The endotracheal tube also provides a pathway for pathogenic oro/nasopharyngeal secretions
to flow directly to the lower respiratory tract, and these can induce respiratory inflammation and
complications such as VAP (Mietto et al., 2013).
VAP describes a ventilator-associated complication focusing on a single infection or pneumonia
affecting a patient receiving invasive mechanical ventilation for a period of greater than, or equal to,
48 hours (American Thoracic & Infectious Diseases Society, 2005; Foglia, Meier, & Elward, 2007).
VAP in both paediatric and adult settings is a constant challenge to the healthcare system. In
developed countries such as the USA, adult VAP incidence rates are reported to be between 0.0 to
5.8 per 1000 ventilator days (Dudeck et al., 2011); however, in developing countries, the rate is
significantly higher at 10 to 41.7 per 1000 ventilator days (Arabi, Al-Shirawi, Memish, & Anzueto,
2008). Recent literature on children suggests that VAP occurs in 1.3% to 36.2% of ventilated children
(Awasthi, Tahazzul, Ambast, Govil, & Jain, 2013; Jordan Garcia et al., 2014; Navoa-Ng et al., 2011;
Patria et al., 2013) with the incidence rates ranging from 0.3 to 31.8 per 1000 ventilator days (Al-
Mousa et al., 2016; Gautam et al., 2012; Rosenthal, Bijie, et al., 2012). VAP is associated with
2
increased duration of mechanical ventilation and subsequent increased length of hospital stay
(Balasubramanian & Tullu, 2014; Gautam et al., 2012; Gupta et al., 2015; Srinivasan, Asselin,
Gildengorin, Wiener-Kronish, & Flori, 2009).
Surveillance of HAIs which can detect the incidence and possible risk factors for VAP is the
cornerstone of every acute care setting’s infection control program (Klompas, Branson, et al., 2014;
Rebmann & Greene, 2010). Surveillance allows hospitals and clinicians to measure the effectiveness
of strategies that are implemented to decrease infection rates (Klompas, Branson, et al., 2014). To be
specific, surveillance is referred to the monitoring and reporting of HAIs events (which are including
VAP and VAE as the focus of interest in this study) as follow:
(1) presence of infection,
(2) the magnitude of the problem and,
(3) the factors that contribute to infections (VICNISS Healthcare Associated Infection
Surveillance, 2015).
A VAP surveillance tool used in adults was first established in 1988 and has undergone serial updates
until 2013 (Mohd Ali, Jauncey-Cooke, & Bogossian, 2019). The VAP surveillance tool relied on
three main criteria:
(1) radiological criteria (chest x-ray);
(2) clinical signs and symptoms;
(3) microbiological laboratory results (Centers for Disease Control and Prevention (CDC) &
National Healthcare Safety Network (NHSN), 2015a; Horan, Andrus, & Dudeck, 2008).
Recently, attention has broadened from VAP to VAE necessitating the development of a more
inclusive surveillance tool incorporating non-infectious ventilator-associated complications; not
limited solely to pneumonia. These complications include pulmonary oedema, barotrauma, acute
respiratory distress syndrome (ARDS) and atelectasis (Klompas, 2013a). The inclusion of a VAE
surveillance tool by the Centers for Disease Control and Preventions (CDC) was partly to address the
limitations of the former VAP surveillance tool and also aimed to improve patient safety by creating
the role of a pay performance indicator (Klompas, 2013b; Muscedere et al., 2013; Septimus, Green,
& Klompas, 2015). VAE is determined using the criteria, which has three ordinal tiers:
(1) ventilator-associated condition (VAC), assessing respiratory deterioration;
(2) infection-related ventilator-associated complication (IVAC), assessing signs of
infection/inflammation;
3
(3) Possible ventilator-associated pneumonia (PVAP), assessing the presence of pathogenic
organisms. The VAE surveillance definition measures VAP as PVAP (Centers for Disease
Control and Prevention (CDC) & National Healthcare Safety Network (NHSN), 2015b).
Given the shift of surveillance tools from VAP to VAE it is important that it’s application in
paediatrics is tested. The surveillance tool developed for children is the PNU1/VAP (Alternative
Criteria for Infants and Children). It relies on similar criteria as described for adults but excludes
microbiological laboratory results (Centers for Disease Control and Prevention (CDC) & National
Healthcare Safety Network (NHSN), 2015a; Horan et al., 2008).
However, the VAP surveillance tools have been criticised because they were found in some studies
to be neither sensitive nor specific sufficiently to detect VAP (Raoof, Baumann, Critical Care
Societies Collaborative, & The Society of Critical Care Medicine, 2014). For example, the
radiological criteria of consolidation may be interpreted as atelectasis or pneumonia or a pleural
effusion. Clinical signs and symptoms (i.e., fever, changes in sputum character, increased suctioning
requirements, worsening cough) are highly subjective and vary from one examiner to another, and
microbiological results may be a consequence of colonisation or infection (Turton, 2008). Hence, the
accuracy of the data derived from these surveillance tools is questionable (Raoof, Baumann, &
Critical Care Society Collaborative, 2014).
Despite this, surveillance data provides important information on VAP preventative strategy
implementation and is an important focus for VAP research (Nair & Niederman, 2015). When VAP
is identified, the treatment relies on the type of organism identified by microbiological sampling
(Kollef, 2011). Treatment of VAP is typically commenced using broad-spectrum antibiotics
considering both patient factors and the pattern of institutional antibiotic resistance (Muscedere et al.,
2008). Ideally, the antibiotic course should start with a broad-spectrum prescription, and be
downscaled after the microbiological results become available (Foglia et al., 2007). This strategy
aims to avoid multi-drug resistant (MDR) strains (Chastre et al., 2003; Nair & Niederman, 2015).
The issue of MDR has become a major threat in intensive care settings and is further hampered by
the decline in production of new drugs (Amin & Deruelle, 2015). However, delays in antibiotic
commencement have been shown to be associated with an increased healthcare burden (Iregui, Ward,
Sherman, Fraser, & Kollef, 2002). Thus, the focus of the VAP/VAE surveillance tools is on
prevention rather than treatment options which aligns with the work of previous scholars such as
Magill et al., (2013) and Magill, Rhodes, and Klompas, (2014).
4
The VAE surveillance tool used in adults also identified most of the events being due to pulmonary
oedema, atelectasis, acute respiratory distress syndrome and pneumonia, which the treatment is then
dependent upon the physiological events identified in each patient (Boyer et al., 2015; Hayashi et al.,
2013; Klein Klouwenberg et al., 2014; Klompas et al., 2011). The potential preventative strategies
for VAE in adults are breathing trials and paired daily spontaneous wakening, early exercise and
mobility, low tidal volume ventilation, fluid management and conservative blood transfusion
thresholds (Klompas, 2015). One study has applied this surveillance definition to paediatrics
(Phongjitsiri et al., 2015). The authors found that VAE was most commonly due to atelectasis, sepsis,
ARDS and shock. This surveillance tool also targets constructing preventative strategies that aim to
improve patient outcomes (Hayashi et al., 2013; Klompas, Magill, et al., 2012). Yet, no study to date
has examined the preventative strategies of VAE in paediatrics which increased the interest in this
study.
Preventative strategies for VAP often can be addressed by the education of staff. Education on VAP
preventative strategies (commonly known as the ‘Ventilator bundle’ or ‘VAP bundle’) and strict
compliance monitoring are crucial in minimising VAP occurrence (Gupta et al., 2014; Jansson,
Kääriäinen, & Kyngäs, 2013; Rello et al., 2002; Smiddy, O' Connell, & Creedon, 2015). VAP
preventative strategies need to be regularly updated and tailored to specific populations with
understanding of the disease emergence and prevalence (Hellyer, Ewan, Wilson, & Simpson, 2016;
Pittet, 2010). VAP preventative strategies are well established in the adult population and have
consistently proven to reduce VAP rates (Khosim et al., 2015; Lu, Shin, & Ding, 2015; Morris et al.,
2011). In contrast, the preventative strategies for VAP in paediatric patients are considered
inconclusive and variable, requiring substantial further research (Brierley, Highe, Hines, & Dixon,
2012; Cooper & Haut, 2013). As a consequence of inadequate paediatric data, the adult VAP
preventative strategies are often adapted to paediatric settings (Cooper & Haut, 2013).
For adult patients, the VAP preventative strategies consist of four to six components:
(1) head of the bed (HOB) elevation greater than 30 degrees;
(2) daily sedation break and daily assessment of readiness to extubate;
(3) the use of gastrointestinal (GI) prophylaxis;
(4) the use of deep vein thrombosis (DVT) prophylaxis;
(5) oral decontamination with chlorhexidine 0.12% and
(6) continuous aspiration of subglottic drainage (Eom et al., 2014; Resar, Griffin, Haraden, &
Nolan, 2014; Wip & Napolitano, 2009).
5
Not all of these preventive strategies suit the paediatric population. Some strategies, such as the daily
sedation break and daily assessment of readiness to extubate may increase the chances of unplanned
extubations and reintubations in infants and young paediatric patients (Klompas, Branson, et al.,
2014). The lack of updated paediatric specific VAP prevention strategies in PICUs suggests there are
grounds for further research to inform paediatric evidence-based VAP preventative strategies. Scant
evidence exists around preventative strategies specifically for VAE in PICU patients (Cocoros,
Priebe, Gray, et al., 2017; Phongjitsiri et al., 2015).
Despite the fact that compliance with VAP preventative strategies (the ventilator bundle) is crucial to
minimise VAP occurrences, modest compliance in implementing the VAP preventive strategies
among healthcare workers has been reported (Cooper & Haut, 2013; Nair & Niederman, 2015); the
prevention of HAIs such as VAP therefore requires education of healthcare workers (Klompas,
Branson, et al., 2014). Several studies have demonstrated improved compliance through education
(Flodgren et al., 2013; Jansson et al., 2013). In addition to the concern for healthcare workers’
compliance, evidence describing the role of PICU parents in HAI minimisation through simple
measures such as hand hygiene is currently also lacking.
Aim
Therefore, this study aims to:
• test the use of the VAE surveillance tool in paediatric patients,
• determine the influence of the updated VAP education focused on specific paediatric VAP
preventative strategies for staff and parents in the PICU, and
• assess the impact following implementation of updated VAP education on VAP and VAE
rates, and VAP preventative strategy compliance.
Research questions
This study addresses four research questions:
1. What is the baseline status of VAP and VAE in paediatric patients as defined by the
PNU1/VAP surveillance and the VAE surveillance tools?
2. What are the current VAP preventative strategies in the PICU?
3. What are the perceptions of parents and nurses in the PICU on the ‘Speaking up for hand
hygiene’ component of the VAP education?
4. Does VAP education and compliance auditing of the preventative strategies with feedback,
reduce VAP and VAE incidence and improve compliance of the VAP preventative strategies?
6
Significance of the study
Reliable VAP and VAE surveillance tool is important to capture the incidence and risk factors that
could be a benchmark for preventative strategies planning in PICU. There is currently no reliable
surveillance tool available to provide precise surveillance data in paediatric patients with ventilator
related complications (Gautam et al., 2012; Klompas, 2010; Srinivasan et al., 2009). Moreover, the
present PNU1/VAP surveillance tool in children is challenging to implement and the effectiveness of
the preventative strategies is difficult to assess (Chang & Schibler, 2016; Cocoros et al., 2016;
Phongjitsiri et al., 2015). Nonetheless, the VAE surveillance tool in adults has shown promising
benefits, and potentially this could be extended to paediatrics (Boyer et al., 2015; Septimus et al.,
2015). With that, both PNU1/VAP surveillance tool and VAE surveillance tool in paediatric need to
be tested.
The significance of the study is link to the concern on continuous challenges in the impact of VAP in
children in PICU. This includes mechanical ventilation-associated complications, lengthened
duration of PICU and hospital stay as well as significant healthcare costs (Chang & Schibler, 2016;
Curley et al., 2006; Foglia et al., 2007; Gautam et al., 2012; Gupta et al., 2015; Srinivasan et al., 2009;
Turton, 2008). Although VAE knowledge in children is not as advanced as our understanding of
VAP, it is estimated that its prevalence ranges from 4.2 to 20.9 episodes per 1000 ventilator days
(Cocoros et al., 2016; Iosifidis et al., 2016; Narayanan, Dixon, Chalkley, Ray, & Brierley, 2016;
Phongjitsiri et al., 2015). Children with VAE have significantly increased hospital morbidity and
mortality (Beardsley, Nitu, Cox, & Benneyworth, 2016; Cocoros et al., 2016; Phongjitsiri et al.,
2015). Given the impact of VAP and VAE on patient safety and healthcare expenditure, these
problems need to be addressed.
Adding to the above, HAIs including VAP are associated with substantial economic and health
burdens which lead to surveillance activities becoming compulsory in many of the parts of the world
for minimising the burden (Martin et al., 2013). In the USA, for example, all healthcare facilities are
required to report their VAP surveillance to the National Healthcare Safety Network (NHSN) as one
indicator of quality assessment, affecting subsequent determination of hospital reimbursement
(Stoeppel et al., 2014). Research on the use of benchmarking and public reporting through
surveillance activities in high-income countries (England, Germany, France and the USA) shows that
these practises help to effect substantial organisational change for patient safety (Haustein et al., 2011;
Klompas, 2013a); this has influenced treatment options and preventative strategy development and
implementation (Haustein et al., 2011; Klompas, 2013a).
7
In developing countries such as the Philippines, India and Argentina, VAP surveillance data is
submitted to the International Nosocomial Infection Consortium (INICC) for benchmarking and
evaluation of inpatient quality reporting programs (Rosenthal, Alvarez-Moreno, et al., 2012).
However, in Australia, VAP surveillance is an optional module, possibly due to the challenging and
labour-intensive application of the surveillance tool (Friedman, Russo, & Richards, 2005). Moreover,
surveillance of VAE in paediatric patients in Australia is not undertaken, and implementation of
preventative strategies into clinical practise for the evaluation of quality improvement thus remains
difficult to measure (Magill et al., 2013; Mietto et al., 2013).
Definitions
The key terms used in this thesis are given below.
Ventilator-associated pneumonia (VAP)
VAP refers to pneumonia, which occurs among children who receive invasive mechanical ventilation
via an endotracheal tube for a period of time greater than, or equal to, 48 hours in PICU. This study
uses the criteria of the PNU1/VAP (Alternative Criteria for Infants and Children) surveillance
definition as outlined by the CDC and NHSN (Centers for Disease Control and Prevention (CDC) &
National Healthcare Safety Network (NHSN), 2015a).
Ventilator-associated events (VAEs)
VAEs1 are ventilator-associated events that occur in ventilated patients in the PICU, according to the
criteria outlined by the CDC and NHSN. These events are divided into three ordinal tiers:
Tier 1: Ventilator-associated condition (VAC);
Tier 2: Infection-related ventilator-associated complications (IVAC); and
Tier 3: Possible ventilator-associated pneumonia (PVAP).
VAP preventative strategies
In this study, the VAP preventative strategies consist of seven individual preventative strategies:
(1) Hand hygiene of PICU staff and parents;
(2) Oral hygiene;
1 In this study, the VAE surveillance definition currently used for adults is applied to paediatric patients in PICU (Centers
for Disease Control and Prevention (CDC) & National Healthcare Safety Network (NHSN), 2015b). IVAC and PVAP
will be mentioned specifically; otherwise, in this study VAC represents VAE. The term ‘surveillance definition’ is used
interchangeably with the term ‘surveillance tool’. ‘VAP/Ventilator bundle’ is used interchangeably with ‘VAP
preventative strategies’.
8
(3) Endotracheal suctioning (open suction);
(4) Endotracheal cuff checks;
(5) Head of the bed (HOB) elevation;
(6) Ventilator circuit checks; and
(7) Early initiation of enteral feeding (Queensland Children's Hospital Paediatric Intensive
Care Unit, 2016).
‘Speaking up for hand hygiene’
‘Speaking up for hand hygiene’ is an initiative or campaign to promote patient safety and prevent the
spread of infection (The Joint Commission, 2018). In this study, ‘Speaking up for hand hygiene’ is
delivered through a pamphlet, educating and reminding parents of the importance of performing hand
hygiene and when they should perform it, and a message encouraging parents to check if they are
washing their hands and check whether healthcare workers or visitors have washed their hands.
The flow of the thesis
This thesis consists of eight chapters (refer to Figure 1.1).
• Chapter 1 is an introductory chapter, which includes study aims, significance of the study,
research questions and definitions of key terms.
• Chapter 2 provides a literature review of the VAP and VAE epidemiological data
and surveillance in paediatrics. This chapter outlines the current debate and ongoing issues
regarding VAP and VAE and discusses the relevance of the existing surveillance definitions
in ventilated paediatric patients.
• Chapter 3 provides a literature review of the current preventative strategies for VAP and VAE
in paediatric patients and the challenges for healthcare workers in implementing the VAP
preventative strategies. This chapter also examines the role that parents play, particularly in
the ‘Speaking up for hand hygiene’ initiative. The review also includes the data that VAP
education may assist in the reduction of VAP prevalence.
• Chapter 4 details the methodologies used in this thesis.
• Chapter 5 presents results and discussion of Phase 1: Retrospective study, which provides the
baseline VAP and VAE status in paediatric patients.
• Chapter 6 describes the implementation of the updated VAP education package and presents
the results and discussion related to Phase 2: VAP compliance auditing. This chapter also
includes survey results from parents and nurses about the ‘Speaking up for hand hygiene’
initiative in PICU.
9
• Chapter 7 presents the results and discussion of Phase 3: Prospective study to assess the
change in VAP and VAE prevalence and preventative strategies compliance following
implementation of the updated VAP education program. This chapter also compares the
findings between Phase 1: Retrospective and Phase 3: Prospective study.
• Finally, Chapter 8 provides a general discussion, limitations, implications and
recommendations for future studies.
Chapter 1
Thesis Introduction
Chapter 2
Literature
Review
Epidemiology
of VAP and
VAE in
children
Chapter 3
Literature
Review
Preventative
strategies,
compliance
issues &
VAP
education
Chapter 4
Methodology
Phase 1:
Retrospective
study
Phase 2:
VAP
education,
VAP
compliance
auditing &
surveys
Phase 3:
Prospective
study
Chapter 5
Results and
discussion:
Phase 1:
Retrospective
study
Chapter 6
Results and
discussion:
Phase 2: VAP
education,
VAP
preventative
strategies
compliance
auditing &
surveys
Chapter 7
Results and
discussion:
Phase 3:
Prospective
study
Chapter 8
General discussion, limitations & conclusion
Figure 1.1: Flow of thesis presentation
10
Summary
This chapter has highlighted the significance of VAP and VAE in critically ill paediatric patients on
invasive mechanical ventilation — in particular, the complexities of the surveillance definitions,
implementation of preventative strategies and ongoing compliance challenges. The chapter concluded
with an overview of the thesis.
11
Literature Review: Epidemiology of VAP and VAE in paediatrics
Introduction to the literature review chapters
A literature search was conducted based on the following electronic databases: PubMed, CINAHL,
Science Direct, Cochrane Review library and Cochrane Database of Systematic Reviews and
Medline. The literature is reviewed over two chapters: (1) epidemiology of VAP and VAE in
paediatrics and (2) preventative strategies, compliance issues and VAP education. This chapter aims
to describe the epidemiology of VAP and VAE in paediatrics and identifies gaps in the literature.
Timeline of surveillance of pneumonia (PNU) in healthcare settings
The surveillance of nosocomial pneumonia was initiated by the Centre for Disease Control and
Prevention (CDC) in 1988 and relied on the combination of three criteria: (1) Radiological findings,
(2) Subjective clinical signs and symptoms, and (3) Laboratory data (Garner, Jarvis, Emori, Horan,
& Hughes, 1988). The CDC criteria remained the same until 2003 when VAP was categorised into
early and late onset (Tablan, Anderson, Besser, Bridges, & Hajjeh, 2004). In 2005, The American
Thoracic Society & Infectious Diseases Society of America limited the tool of VAP to pneumonia in
patients on mechanical ventilation for at least 48 hours (American Thoracic & Infectious Diseases
Society, 2005). The CDC further classified pneumonia (PNU) into three categories (PNU1, PNU2,
and PNU3) and provided alternative criteria for infants and children (PNU1/VAP) (see Figure 2.1)
(Horan et al., 2008). The current VAP criteria for infants and children are available in the CDC and
NSHN document ‘Device-associated modules; pneumonia (ventilator-associated [VAP] & non-
ventilator associated pneumonia [PNU]) event’ (Centers for Disease Control and Prevention (CDC)
& National Healthcare Safety Network (NHSN), 2015a). The current tool, however, is frequently
questioned due to its subjective criteria and inter-observer variability between infection control
personnel and assessors (Chang & Schibler, 2016; Hayashi et al., 2013; Klompas, 2010).
In 2013, the CDC released a new surveillance tool to capture events extending beyond VAP. This
new tool is referred to as VAE and it contains three nested tiers:
(1) Tier 1 is the ventilator-associated condition (VAC);
(2) Tier 2 is the infection-related ventilator-associated complication (IVAC); and
(3) Tier 3 is the possible ventilator-associated pneumonia (PVAP) (Centers for Disease
Control and Prevention (CDC) & National Healthcare Safety Network (NHSN), 2015b).
A diagnosis of VAE can be made when the patient’s physiological or mechanical ventilation
parameters meet the first-tier criteria. If the first-tier (Ventilator associated condition (VAC) criteria
12
are met, the second tier (Infection-related ventilator-associated complications (IVAC)) is examined.
If the second-tier criteria are met, the third tier (Possible ventilator-associated pneumonia (PVAP) is
examined. The timeline for surveillance of pneumonia, with its complex and changing criteria, is
illustrated in Figure 2.1.
13
Figure 2:1: Timeline for surveillance of pneumonia (PNU) in a healthcare setting
7 10 13 16 19 22 25 28 31
7 10 13 16 19 22 25 28 31
CDC definition for
nosocomial
infection
1988
Differentiated
pneumonia into two age
categories:
Patient ≤ 12 months and
> 12 months using:
Combination of clinical,
radiological &
laboratory findings
CDC Guidelines for
preventing
healthcare-associated
pneumonia
2003
VAP classified into either
early-onset (develops
within 96 hours of patients’
admission to an ICU or
intubation for mechanical
ventilation); late onset
(develops after 96 hours of
patients’ admission to an
ICU or intubation for
mechanical ventilation)
2005
CDC Guidelines
published for the
management of adults
with hospital acquired,
ventilator-associated &
healthcare-associated
pneumonia
Defined VAP as
pneumonia in
mechanical ventilation in patients with
mechanical ventilation
of at least 48 hours CDC/NSHN surveillance
tool of healthcare
associated infection &
criteria for specific types
of infections in acute care
settings
Revised in 2002 &
widely used in 2008
Replaced term “nosocomial” to healthcare-associated infection”
(HAI) and classified Pneumonia
into three categories: 1. PNU1 (Clinical Defined
Pneumonia)
i) For any patient
ii) Alternate criteria for infant ≤ 1
year old & for child > 1 year old
or ≤ 12 years old
2. PNU2 (Pneumonia with
Specific Laboratory Findings) 3. PNU3 (Pneumonia in
Immunocompromised Patients)
2013 until present
Ventilator-associated
Events (VAE) -
Surveillance/ protocol
mandated to adult
patient population only
Retained the surveillance tool
for children as stated in the
2008 guideline: available in a
document titled as Device-
associated module;
pneumonia (Ventilator-
associated (VAP) & non-
ventilator associated
Pneumonia (PNEU) Event
History of surveillance of pneumonia in healthcare settings
Ventilator associated
consist of three tiers:
1. VAC (Ventilator
associated condition)
2. IVAC (Infection-
related ventilator-
associated
complications)
3. PVAP (Possible
ventilator-associated
pneumonia)
Minor changes for PNU 1 (Clinical Defined
Pneumonia) For ii) alternate criteria for
child >1 year old or ≤ 12
years old.
- Fever (>38.0 oC) or
hypothermia (<36.0 oC)
14
VAP and VAE: Pathogenesis and factors for acquiring VAP/VAE
The development of VAP is predominantly related to oro-digestive tract colonisation and pathogenic
bacteria aspiration from invasive devices such as the endotracheal tube (Coffin et al., 2008; Foglia et
al., 2007; Kollef, 2004; Safdar, Crnich, & Maki, 2005; Tablan et al., 2004; Zolfaghari & Wyncoll,
2011). Additionally, ventilator circuits and suction equipment are associated with pathogenic bacteria
aspiration (Safdar et al., 2005; Tablan et al., 2004). This theory is supported by the microbiological
examination of the endotracheal secretions of VAP patients which is similar to the organisms found
in the naso-oropharyngeal and gastric secretions (Chastre & Fagon, 2002). The common organisms
for VAP are Staphylococcus aureus, Klebsiella pneumonia, Acinetobacter baumannii and
Enterobacter species (Cooper & Haut, 2013; Sachdev et al., 2013). Factors associated with the
development of VAP are (1) aspiration of pathogenic secretions, (2) colonisation of the oro-digestive
tract and (3) use of contaminated equipment (Coffin et al., 2008; Hellyer et al., 2016).
VAE as a broader phenomenon extends beyond a pneumonic process. VAE identifies infectious and
non-infectious ventilator-associated complications based on the three nested tiers. The first tier,
referred to as VAC, mainly focuses on respiratory and non-respiratory complications or conditions
by evaluating the deterioration of oxygenation parameters — thus the patient requires a sustained
increase in ventilatory support. The second tier is the infection-related IVAC that emerges from the
first-tier of the VAC and further defines whether VAC is due to an infection or not. This is
accomplished by adding another two criteria: abnormal body temperature or abnormal white blood
cell (WBC) count, and the initiation of new antibiotics for four days or more. The final tier is called
PVAP, which is a subset of IVAC that may indeed be pneumonia. In these circumstances, positive
microbiological findings from the respiratory secretions with specific thresholds are required (Figure
2.2) (Centers for Disease Control and Prevention (CDC) & National Healthcare Safety Network
(NHSN), 2015b; Goutier et al., 2014). Thus, the pathogenesis of the VAC, IVAC and PVAP relate to
the type of complications or conditions involved (Klompas, 2015).
15
Figure 2:2: Ventilator-associated events in the respective tiers
The global incidence and the impact of VAP and VAE
The incidence of VAP among paediatric patients varies across the globe. The incidence in paediatrics
is reported to be as high as 31.8 per 1000 ventilator days (Rasslan et al., 2012). A smaller study in
2002 in the USA reported the rate of 11.6 per 1000 ventilation days involving 34 episodes of
mechanical ventilation in 30 patients (Elward, Warren, & Fraser, 2002). More than 10 years later, in
2015, the VAP rate in a multicentred PICU study in the USA reported a rate of 7.1 per 1000 ventilator
days (Gupta et al., 2015).
In Europe, the VAP rate in a single PICU in Greece was found to be 15.3 per 1000 ventilator days
with an incidence density of 24.4% (30/127 patients) in a one-year retrospective study (Stabouli et
al., 2012). A retrospective study in the United Kingdom in 2012 revealed that the incidence was 9.2
per 1000 ventilator days (Ismail, Darbyshire, & Thorburn, 2012). A six month prospective study in
the United Kingdom reported a VAP rate of 2.4 per 1000 ventilator days (4/325 patients) with a 1.7%
incidence density (Narayanan et al., 2016) compared to a prospective study by Patria et al. (2013),
which found that the incidence density in Italy (30/451 patients) was 6.6%.
In Japan, Hatachi, Tachibana, and Takeuchi (2015) retrospectively examined 426 patients during the
2013 calendar year, reporting a VAP rate of 3.5 per 1000 ventilator days, and an incidence density of
1.2% (5/426 patients). In India, the incidence density of VAP was substantially higher; 36.2% (38/105
VAC
• Identification of respiratory and non-respiratory complications based on increased requirement in ventilatory support
IVAC• Further identification of VAC;
with or without an identified infectious source
PVAP • Identification of IVAC that may indeed be pneumonia
16
patients) (Awasthi et al., 2013). In Australia, a study in 2010 reported an incidence density of 6.7%
(18/269 patients) with a VAP rate of 7.02 per 1000 ventilator days (Gautam et al., 2012).
A few paediatric studies have evaluated epidemiological data across multi-centered PICUs. One study
in the USA collected data from 2007–2012 in 64 PICUs (Patrick et al., 2014). This study revealed
that the VAP rate declined over time from 1.9 to 0.7 per 1000 ventilator days (Patrick et al., 2014). A
multi-centered prospective study across six PICUs (300 patients) in Spain reported a VAP rate of 9.4
per 1000 ventilator days (Jordan Garcia et al., 2014). A prospective study involving 16 PICUs (2081
patients) in the USA revealed the VAP rate was 7.1 per 1000 ventilator days (Gupta et al., 2015). In
a study that reported VAP data under the International Nosocomial Infection Consortium (INICC)
involving eight PICUs, the VAP rate was 8.1 per 1000 ventilator days (Rosenthal, Alvarez-Moreno,
et al., 2012). This study was conducted in five developing countries, including Columbia, El-
Salvador, India, the Philippines and Turkey, in 2012.
The implication of high incidence rates on critically ill children contributes to a burden on the
healthcare system. The burden of VAP is primarily due to the increased duration of mechanical
ventilation. Gautam et al. (2012), reported that a child with VAP spends seven extra days on a
ventilator compared to those without VAP (11.9 days versus 4.9 days). Balasubramanian and Tullu
(2014) reported similar findings; 22 days versus five median days, p < 0.001 for a patient with VAP
and without VAP respectively. Gupta et al. (2015), found that children with VAP spend 11 days on
ventilator in the PICU versus three median days, p < 0.001 in patient without VAP.
VAP also contributes to a significant increase in the length of PICU stays (mean 46.0 ± SD 43.7 days
versus mean 9.1- ± SD 9.3 days, p< 0.001) in patients with VAP compared to those without VAP
(Stabouli et al., 2012). A prolonged PICU stay correlates with higher hospital costs (Elward et al.,
2002; Foglia et al., 2007). The cost of VAP has been calculated at US$308,534 compared to
US$252,652 in those patients without VAP (Srinivasan et al., 2009). Brilli et al. (2008), reported that
the extra hospital costs for VAP patients were US$156,110 compared with US$104,953 for non-VAP
patients, or approximately US$50,000 in additional hospital costs per ventilator episode. Utilisation
of PICU beds (without treatment) costs incurred are as high as €9207.4 +/-8737.8 versus €1820.0 +/-
1850.5, p < 0.001 in those children with VAP compared to those without VAP (Stabouli et al., 2012).
Such costs pose a significant financial burden on hospitals when a limited reimbursement for hospital
costs is mandated (Agarwal et al., 2010; Restrepo et al., 2010).
17
In some paediatric studies, those with VAP have been shown to have a threefold increased probability
of mortality (p < 0.007) (Gupta et al., 2015). Absolute hospital mortality has been shown to increase
by 8.1% in children with VAP compared to those without (10.5% versus 2.4%) (Srinivasan et al.,
2009). The mortality of VAP patients in a study conducted by Bigham et al. (2009) was 19.1% versus
the non-VAP at 7.2 %, (p= 0.01).
There is currently limited data on the incidence of VAE in children but it is estimated to range from
1.1 to 20.9 per 1000 ventilator days as a result of variations in the surveillance criteria across studies
(Beardsley et al., 2016; Cocoros et al., 2016; Iosifidis et al., 2016; Narayanan et al., 2016; Phongjitsiri
et al., 2015). Two of these five studies report significantly high incidence rates of VAE — 20.9 and
11.2 per 1000 ventilator days (Iosifidis et al., 2016; Phongjitsiri et al., 2015). The remaining three
studies reported incidence rates of 1.1, 2.1 and 4.2 per 1000 ventilator days respectively (Beardsley
et al., 2016; Cocoros et al., 2016; Narayanan et al., 2016). The majority of these studies support the
view that patients with VAE have a significantly increased duration on mechanical ventilation support
and longer PICU and hospital stays (Beardsley et al., 2016; Cocoros et al., 2016; Iosifidis et al., 2016;
Phongjitsiri et al., 2015).
Overall, recent VAP incidence rates show wide inter- and intra-continent variability, reflecting the
variation in resource distribution, VAP/VAE surveillance and the application of the surveillance tools
(Aelami, Lotfi, & Zingg, 2014; Mourani & Sontag, 2017). Despite this heterogeneity, the incidence
in single PICUs is still considered high and indicates that appropriate attention should be given to
address this threat to patient safety.
Risk factors for VAP and VAE
The study of risk factors for VAP offers the opportunity to improve our understanding of the
likelihood of developing an infection, enabling targeted direct interventions (Cook & Kollef, 1998;
Gautam et al., 2012; Kollef, 2004). Various factors may contribute to the development of VAP in
paediatric patients, including age-related physiological differences and comorbidities
(Balasubramanian & Tullu, 2014; Mourani & Sontag, 2017), but a limited number of specific
paediatric studies evaluate risk factors for VAP (Liu et al., 2013; Srinivasan et al., 2009).
Table 2.1 illustrates the paediatric studies published in the last 10 years (2005–2015) that focus on
the risk factors associated with VAP development. The majority of the studies were conducted in the
USA and in single PICU centers (Bigham et al., 2009; Srinivasan et al., 2009), with only one
multicentered study (Gupta et al., 2015). Two studies were conducted in Brazil (Casado, de Mello,
18
de Aragão, de Albuquerque, & Correia, 2011; Kusahara, Enz Cda, Avelar, Peterlini, & Pedreira Mda,
2014) and one study each in India, Australia and the Netherlands (Awasthi et al., 2013; Gautam et
al., 2012; Roeleveld et al., 2011). The risk factors reported vary across studies, with reintubation the
highest reported risk factor for the development of VAP (Awasthi et al., 2013; Gautam et al., 2012;
Gupta et al., 2015; Kusahara et al., 2014; Srinivasan et al., 2009). This is also one of the risk factors
identified in a systematic review by Liu et al. (2013). The second most reported risk factor is
microaspiration secondary to reduced tolerance of enteral feeding (Casado et al., 2011; Kusahara et
al., 2014; Srinivasan et al., 2009). These findings suggest that the modifiable risk factors of VAP
outweigh the non-modifiable risk factors such as subglottic/tracheal stenosis, trauma, female gender,
post-surgical admission diagnosis and paediatric risk index of mortality version 3 score (PIMS 3)
(Bigham et al., 2009; Roeleveld et al., 2011; Srinivasan et al., 2009).
To date, three studies have examined the risk factors for VAE in children. According to Phongjitsiri
et al. (2015) the risk factors for VAE among paediatric patients include immunocompromised status,
chronic respiratory disease and tracheostomy dependence. Cocoros, Priebe, Gray, et al. (2017),
identified, through a nested case-control study, neuromuscular blockade (OR, 2.29; 95% CI, 1.08–
4.87), positive fluid balance (OR, 7.76; 95% CI, 2.10–28.6), and blood product use (OR, 1.52; 95%
CI, 0.70–3.28) as risk factors for VAE. Beardsley et al. (2016), reported trauma as an independent
risk factor based on multivariate analysis (adjusted OR= 3.10%, CI= 1.15-8.38), but this study was
limited only to children who had positive respiratory cultures.
In Australia, there has been only one study of a PICU during the past 10 years, and it only examined
epidemiological data for VAP (Gautam et al., 2012). The lack of evidence regarding the risk factors
for VAE has significant implications for the development of preventive strategies (Cocoros et al.,
2016; Liu et al., 2013; Phongjitsiri et al., 2015; Srinivasan et al., 2009) and the effectiveness of
existing interventions in reducing the incidence of VAP/VAE is not able to be reliably determined.
Hence there is an urgent need to measure the potential risk factors for VAP and VAE to inform the
planning of preventative strategies.
19
Table 2.1: Risk factors for VAP in children: Studies published from 2005 to 2015
No Authors Country Single/
Multicentred
Risk factors of VAP
1 Bigham et al. (2009) USA Single Subglottic/tracheal stenosis, trauma, and tracheostomy
2 Srinivasan et al. (2009) USA Single Univariate analysis: the use of metoclopromide, reintubation, worsening oxygenation
(measured by partial pressure of oxygen (PaO2/ FiO2 ratio), recipient of blood products and
the presence of enteral feeding.
Multivariate analysis; female gender, postsurgical admission diagnosis, the presence of
enteral feeding, and use of narcotic medications.
3 Roeleveld et al. (2011) Netherland Single PIMS 3 and transfusion of fresh frozen plasma
4 Casado et al. (2011) Brazil Single Longer stay on ventilation (OR), 1.04; 95% (CI), 1.01–1.08), use of gastric tube (OR, 2.88;
95% CI, 1.41–5.87), and of sedatives/analgesics (OR, 2.45; 95% CI, 1.27–4.72)
5 Gautam et al. (2012) Australia Single Univariate analysis: reintubation, the absence of tube feeding and absence of stress ulcer
prophylaxis.
Multivariate analysis: tube feeding (HR), 0.27; 95% CI, 0.09–0.85; p=0.02) and stress ulcer
prophylaxis (HR), 0.29; 95% CI, 0.11–0.76; p=0.01) reduced the incidence of VAP
6 Awasthi et al. (2013) India Single Reintubation within 72 hours of extubation
7 Kusahara et al. (2014) Brazil Single The use of nasogastric tubes (OR, 5.278; p<0.001), intermittent administration of nutritional
formula (OR, 6.632; p=0.005), emergency reintubation (OR, 2.700; p=0 .02), use of
vasoactive drugs (OR, 5.108; p=0.009), duration of mechanical ventilation (p< 0.001)
8 Gupta et al. (2015) USA Multicentred (16
PICUs)
Reintubation and ‘part-time’ ventilation
Notes: PIMS 3: paediatric risk index of mortality score 3; OR: odds ratio; CI: confidence interval; HR: hazard ratio
20
Diagnostic criteria for pneumonia (PNU/VAP)
According to the CDC and NHSN, there are three categories of ventilator associated pneumonia:
PNU1, PNU2 and PNU3. Each category requires two to three criteria to be met, for a diagnosis of
VAP.
• The first criterion is radiological: this requires two or more serial imaging test results (with
underlying cardiac or respiratory disease) or one or more imaging test results (without
underlying cardiac or respiratory disease). The imaging test requires at least one of the
following: new and persistent or progressive and persistent: (a) consolidation, (b) infiltration,
(c) cavitation and pneumatoceles.
• The second criterion is clinical: i.e., clinical signs and symptoms characterised by an abnormal
WBC count; body temperature instability; abnormal respiratory signs and symptoms (such as
changes in sputum character or increased suctioning requirements); new onset or worsening
cough; dyspnoea or tachypnoea or apnoea; tachycardia or bradycardia; rales or bronchial
breath sounds; wheezing or rhonchi; worsening oxygenation.
• The final criterion is based on laboratory results with or without positive microbiological
findings (Centers for Disease Control and Prevention (CDC) & National Healthcare Safety
Network (NHSN), 2015a; Horan et al., 2008).
The CDC alternative criteria for infants and children (PNU1)
To addresses the physiological differences in paediatrics the CDC has developed alternative criteria
for a diagnosis of VAP.
PNU1 in adults is ‘clinically defined pneumonia’, a diagnosis that requires radiological criteria and
the signs and symptoms but without an identified pathogen. The paediatric PNU1 (alternative criteria
for infants and children) is divided into two age categories: infants ≤ 1 year old and children > 1 or ≤
12 years (Table 2.2) (Centers for Disease Control and Prevention (CDC), 2015). PNU2 is pneumonia
with common bacterial or filamentous fungal pathogens and has specific laboratory findings or
pneumonia with viral, Legionella and other bacterial pneumonia with definitive laboratory findings.
PNU3 is pneumonia in immuno-compromised patients. PNU2 and PNU3 requires all three criteria
regardless of the age of the patients (Centers for Disease Control and Prevention (CDC) & National
Healthcare Safety Network (NHSN), 2015a; Horan et al., 2008).
The PNU1 (alternative criteria for infants and children) tool remain the main surveillance tool for
VAP in children (Centers for Disease Control and Prevention (CDC) & National Healthcare Safety
Network (NHSN), 2015a; Horan et al., 2008). However, the tool is highly subjective, correlates
21
poorly with histology, is neither sensitive nor specific and it is time consuming when using the tool
because complexities of criteria involved, making it difficult to conduct the surveillance (Klompas,
2010; Klompas et al., 2011; Venkatachalam, Hendley, & Willson, 2011).
Moreover, the exclusive focus on VAP as the main ventilator-associated complication fails to
acknowledge that there are other potential complications related to mechanical ventilation (Klompas,
2013a, 2013b; Klompas, Kleinman, & Murphy, 2014).
Diagnostic criteria for VAE surveillance
VAE is a new paradigm that views ventilator-associated complications from a different and broader
perspective (Muscedere et al., 2013; Septimus et al., 2015). The diagnostic criteria have been divided
into three tiers. The requirements for each tier are as follows:
• Tier 1: VAC — requires the objective data derived from ventilator settings; fraction of
inspired oxygen (FiO2) or positive end expiratory pressure (PEEP) to evaluate worsening
oxygenation within a certain period of stability;
• Tier 2: IVAC — requires meeting tier 1 (VAC) criteria plus demonstrate infection indicators;
body temperature or WBC count, inclusive of the antibiotic initiation and continued for at
least four days;
• Tier 3: PVAP — requires meeting tier 1 (VAC) and tier 2 (IVAC) criteria plus confirmed
microbiological findings from the respiratory secretions (Centers for Disease Control and
Prevention (CDC) & National Healthcare Safety Network (NHSN), 2015b).
A summary of the VAE tiers with their respective requirements and criteria is given in Table 2.3.
22
Table 2.2: CDC PNU1/VAP Tool: Alternative Criteria for Infants and Children
PNU1 VAP Tool Radiological criteria
For both aged categories a) Patient with underlying diseases has two or more imaging test results with one of the following; b)
Patient without underlying diseases has 1 or more imaging test results with one of the following:
New and persistent OR progressive and persistent:
1. Infiltrate
2. Consolidation
3. Cavitation
Pneumatoceles, in ≤1 year old
Signs and symptoms
Alternate criteria for infant ≤ 1
year old
Worsening gas exchange (e.g., oxygen desaturations)
[e.g., pulse oximetry <94%], ↑ oxygen requirement or ↑ ventilation demand) and three of the following:
▪ Temperature instability
▪ Leukopenia (≤4,000 WBC/mm3) or leucocytosis (≥15,000 WBC/mm3) and left shift (≥10%
band forms)
▪ New onset of purulent sputum, or change in the character of the sputum, or ↑respiratory
secretions, or ↑ suctioning requirements
▪ Apnoea, tachypnoea, nasal flaring with retraction of the chest wall or grunting
▪ Wheezing, rales, or rhonchi
▪ Cough
▪ Bradycardia (<100 beats/min) or tachycardia (>170 beats/min)
Alternate criteria for Children >1
or ≤ 12 year old
At least three of the following:
▪ Fever (>38.0°C/100.4°F) or hypothermia (<36.0°C/96.8°F)
▪ Leukopenia (≤4,000 WBC/mm3) or leucocytosis (≥15,000 WBC/mm3)
▪ New onset of purulent sputum, or change in the character of the sputum, or ↑ respiratory
secretions, or ↑ suctioning requirements
▪ New onset of worsening cough, or dyspnoea, apnoea, or tachypnoea
▪ Rales or bronchial breath sounds
▪ Worsening gas exchange (e.g., oxygen desaturations [e.g., pulse oximetry <94%]↑ oxygen requirement
or ↑ ventilation demand)
23
Table 2.3: The CDC VAE tiers with respective requirements and criteria
VAE Tier Requirements and criteria
1. Ventilator associated condition
(VAC)
Pre-requirement: The patient is required to have a baseline period of stability or improvement on
ventilator for ≥ 2 calendar days of stable or decreasing *daily minimum FiO2 or PEEP values.
VAC criteria:
1. Increase FiO2 ≥ 0.20 OR PEEP ≥3 cmH2O
2. Sustained for 2 days
*Daily minimum defined by the lowest value of FiO2 or PEEP during a calendar day that is maintained
for at least 1 hour.
2. Infection- related ventilator
associated complications (IVAC)
Pre-requirement: The patient is required to meet the VAC criteria
IVAC criteria:
1. Temperatures <36oC or >38oC OR abnormal white blood cell (WBC) count (≤ 4,000 cells/mm3 or ≥
12, 000 cells/mm3)
AND
2. New antimicrobial agent (s) is started and continued for ≥ 4 days within 2 days of the increase in PEEP
or FiO2.
3. Possible ventilator associated
pneumonia (PVAP)
Pre-requirement: The patient is required to meet the IVAC criteria
PVAP criteria:
1. Positive culture of respiratory secretion (via endotracheal aspirate, bronchoalveolar lavage (BAL), lung
tissue, or protected specimen brush)-met the quantitative or semi-quantitative thresholds OR
2. Purulent respiratory secretions and positive culture via specimens in criteria 1, but not meeting those
thresholds for growth OR
3. One positive test from the pleural fluid or lung histopathology or diagnostic test for Legionella species
or respiratory secretion positive for the viral organism, within 2 calendar days of meeting the IVAC
criteria
Notes: Purulent secretions=≥25 polymorphonuclear cells (PMNs), ≤10 squamous epithelial cells per low-powered field (LPF); Positive
culture=endotracheal aspirate ≥105 colony forming units (CFU)/ml
24
Ventilator associated events (VAE) in children: A review of literature
Publication
A literature review of ventilator-associated events in children was undertaken by this author and
supervisors and was published in Australian Critical Care (see below). The focus was on the
application of the new VAE surveillance tool in the paediatric population and the potential challenges
in clinical practise. This chapter is presented mostly as per the published version, but some
information is presented in other chapters, such as Chapter 1, Chapter 2 (sub-sections), and figures
and tables within this thesis. The page numbering has been adjusted to fit the overall thesis style, and
references listed will be at the end of the thesis.
Publication
Mohd Ali, N. A., Jauncey-Cooke, J., & Bogossian, F. (2019). Ventilator-associated events in
children: A review of literature. Australian Critical Care, 32(1), 55-62.
doi:10.1016/j.aucc.2018.11.063
ABSTRACT
Background: The complexity and variation in ventilator associated pneumonia (VAP) tools in
paediatrics may pose threats to the reliable identification of VAP. The revision of the surveillance
tool to ventilator-associated event (VAE) has been mandated in adult populations to overcome these
issues. However, the evidence for application of this tool is unknown in children.
Objectives: To review the evidence on the application of the new VAE surveillance tool in the
paediatric population and examine the potential challenges in clinical practise.
Review methods: A systematic approach was used to locate and synthesise the relevant paediatric
literature. Studies were appraised according to epidemiological appraisal instrument (EAI) and the
grades of evidence in the National Health Medical Research Council (NHMRC) guidelines.
Results: Seven studies met the inclusion criteria. Quality of study methods was above 50% on the
EAI. The overall grade of evidence was assessed as C (satisfactory). The incidence of VAE in
children ranged from 1.1 to 20.9 per 1000 ventilator days as a result of variations in surveillance
criteria across included studies. There is little agreement between the new VAE and PNU/VAP
surveillance tool in the identification of VAP. Challenges in the application of VAE surveillance were
related to: the difference in modes of ventilation used in children versus adults; inconclusive criteria
25
tailored to paediatric samples; and a lack of data. The latter problem provides support for automatic
data extraction to be applied in paediatric studies.
Conclusion: This review demonstrated promising evidence using the new VAE surveillance tool to
define the VAE in children, but the level of the evidence is low. Before the possibility of real
implementation in clinical settings, challenges related to VAE paediatric specific criteria and the
value of automated data collection need to be considered.
AIMS/OBJECTIVES
This literature review was conducted to synthesise the available literature on VAE to answer the
following research questions:
1. Is the VAE surveillance tool used in adults able to identify ventilator-associated complications in
the paediatric population?
2. What are the potential challenges in the application of the new VAE surveillance tool in paediatric
clinical practise?
MATERIAL AND METHODS
A systematic literature search was conducted using the following electronic databases: PubMed,
CINAHL, ScienceDirect, Cochrane Review library and Cochrane Database of Systematic Reviews,
and MEDLINE. Subject headings (MeSH or CINAHL headings) were included as follows:
‘pneumonia, ventilator-associated’, ‘complication, ventilator-associated’, ‘surveillance’, and some
minor/subheadings. Key search terms were ‘ventilator associated pneumonia’, ‘ventilator associated
event’, ‘child*’, ‘criteria’, ‘surveillance’, ‘p(a)ediatric’, ‘pneumonia’, ‘intensive care unit’. The
Boolean operators OR, AND, and NOT were applied. The wildcard symbols were not applied in
Google scholar searches. The search considered all relevant literature related to VAEs (as per the new
surveillance tool) and was limited to English language publications from January 2010 to February
2018.
The reference lists from identified articles were also hand searched to reduce the possibility of
excluding relevant literature. To maximise the effectiveness of search strategies, the literature search
was undertaken in consultation with an expert health librarian. Screening by the title and abstract was
carried out by one author if this was insufficient to make a decision on the article, the full text was
obtained. The inclusion of retrieved publications was reviewed and agreed by two authors, and there
was no disagreement. The predefined inclusion criteria were the following: (1) paediatric patients
aged 0 to18 years and (2) received invasive mechanical ventilation.
26
A purpose-built data extraction form consisted of study setting, aims, participant population, and
outcome measure. One author extracted data from the included articles. The data were then evaluated
by the first author to obtain the methodological quality using an epidemiological appraisal instrument
(EAI) described by Genaidy et al. (2007). There were 39 of 43 items in the EAI applicable for the
purpose of review. Four items that were not applicable were the following: (1) adverse effects
reported that may be consequences of the intervention, (2) newly incident cases consideration
‘(applied only in case-control design)’ (3) subjects randomisation, and (4) randomisation assignments
concealment. The scoring for the items was either “yes” (2 points), “partial” (1 point), “no” (0 point),
or “unable to determine” (0 point) (Genaidy et al., 2007). Cut-off points between high and low
quartiles were based on EAI scores, (more than 50% as higher quality and less than 50% as low
quality) as described in a study conducted by Nix, Smith, and Vicenzino (2010). Discussions were
held, and consensus was obtained for any disagreements with the other two authors. Finally, grading
was undertaken using the NHMRC Description of Evidence Levels, Grade of Recommendation and
Body of Evidence Assessment Matrix (The National Health and Medical Research Council
(NHMRC), 2000).
SEARCH RESULTS
The search retrieved a total of 328 potential articles during the initial stage of screening. The authors
then excluded publications that did not use the new VAE surveillance tool, duplicates, and those
studies undertaken in adults and animal models. A total of seven articles remained for inclusion in
this review, which was conducted in accordance with the Preferred Reporting Items for Systematic
Reviews and Meta-Analyses (Figure 2.3) (Moher, Liberati, Tetzlaff, Altman, & The PRISMA Group,
2009). Two included studies were conducted in Europe (Iosifidis et al., 2016; Narayanan et al., 2016)
and the remaining included studies were from the USA (Beardsley et al., 2016; Cirulis, Hamele,
Stockmann, Bennett, & Bratton, 2016; Cocoros et al., 2016; Phongjitsiri et al., 2015; Taylor,
Noronha, Wichman, & Varman, 2014). Six included studies were retrospective in design, and one
was a prospective study (Narayanan et al., 2016). Overall, the level of evidence was level III-3
prognosis with a satisfactory body of evidence (Grade C) based on The National Health and Medical
Research Council (NHMRC) guidelines (The National Health and Medical Research Council
(NHMRC), 2000). The summaries of the individual study are provided in Table 2.4. The quality of
most of the studies was more than 50% (high) (Beardsley et al., 2016; Cirulis et al., 2016; Cocoros et
al., 2016; Iosifidis et al., 2016; Phongjitsiri et al., 2015; Taylor et al., 2014) with only one study
reported of low quality (less than 50%) (Narayanan et al., 2016) on the EAI. In all studies, the study
design and source of the population are clearly described.
27
Figure 2:3: PRISMA flow diagram of literature selection
Records identified through
database searching
(n=478)
Screen
ing
Inclu
ded
E
ligib
ilit
y
Iden
tifi
cati
on
Additional records
identified through other
sources
(n=1)
Records after duplicates removed
(n=151)
Records screened
(n=328)
Records excluded by
title and abstract
(n=182)
Full-text articles
assessed for eligibility
(n=146)
Full-text articles
excluded, with reasons
(n=139)
Adult=36
Animal=4
Did not include/use the
VAE surveillance tool=99
Studies included in the
review
(n=7)
28
Table 2.4: Study summaries and agreement between surveillance tools
Authors Location Study
design/population
NHMRC Level of
evidence
Incidence or prevalence of VAE
Agreement between surveillance
tools
Phongjitsiri et al.
(2015)
USA Retrospective cohort
(606 patients)
III-2 aetiology
evidence
VAC=20.9/1000 ventilator days.
IVAC=12.9/1000 ventilator days.
PVAP=7.1/1000 ventilator days.
Probable VAP (PrVAP)=3.7/1000
ventilator days, 16.3% Undetermined
infection=2.1 /1000 ventilator days.
New VAE versus PNU/VAP=41
patients versus 9 patients
Cocoros et al.
(2016)
USA Multicentre
Retrospective cohort
(8862 patients)
III-3 prognosis
evidence
VAC=1.1-4.6/1000 ventilator days
depending to ICU types.
No comparison of tools was carried
out, however, attempted to test
different thresholds of VAC -
proposed Fi02 of 0.25 & mean
arterial pressure (MAP) at 4 for
paediatric patients
Narayanan et al.
(2016)
United
Kingdom
Prospective cohort
(258 patients)
III-3 prognosis
evidence
VAC= 4.2/1000 ventilator days.
PVAP=1.8/1000 ventilator days.
New VAE versus PNU/VAP=7
patients versus 4 patients
Beardsley et al.
(2016)
USA Retrospective cohort
(217 patients)
III-3 prognosis
evidence
IVAC=2.16/1000 ventilator days.
New VAE versus PNU/VAP=4
mechanical ventilation episodes
versus 5 mechanical ventilation
episodes
Only 1 mechanical ventilation met
both tool
Cirulis et al.
(2016)
USA Retrospective cohort
(119 patients)
III-3 prognosis
evidence
9 patients met the new VAE surveillance
tool.
22 patients identified using the modified
VAE surveillance tool.
New VAE & Modified New VAE=
versus PNU/VAP=Poor sensitivity
but good specificity
Iosifidis et al.
(2016)
Greece Retrospective cohort
(119 patients)
III-3 prognosis
evidence
11.2/1000 ventilator days. New VAE versus PNU/VAP=12
patients versus 13 patients
Agreement= Poor agreement (5
patients met both surveillance tools)
Taylor et al.
(2014)
USA Retrospective cohort
(285 patients)
III-3 prognosis
evidence
17 patients experienced PVAP. New VAE versus PNU/VAP =17
patients versus 15 (9 patient met
both surveillance tools)
29
RESULTS
Is the new VAE surveillance tool used in adults able to identify ventilator-associated
complications in the paediatric population?
Four studies adopted the new VAE surveillance tool in paediatric patients to describe the prevalence
of VAE in their respective units. Iosifidis et al. (2016) conducted a retrospective study to evaluate the
new surveillance tool for VAC, IVAC and PVAP and compared this to an earlier CDC VAP tool
referred to as PNU1/VAP (Table 2.2). The study, undertaken in 2011, involved 119 children admitted
to PICU. It found that 19 patients met the VAC tier with the incidence rate of 11.2 per 1000 ventilator
days. Of 19 patients meeting the VAC tier, 14 met the IVAC tier and from those 14 cases, 12 met the
PVAP tier. The researchers also evaluated the same cohort of patients using the PNU1/VAP
surveillance tool and the results demonstrate poor agreement between the two surveillance tools,
despite the same incidence reported. Only five patients met the VAP criteria classified by both
surveillance tools.
Phongjitsiri and collegues (2015), using a similar study design with a larger cohort of patients’ records
(n= 606), reported the incidence rate of VAE was 20.9 per 1000 ventilator days (Phongjitsiri et al.,
2015). Of these, the incidence of IVAC, PrVAP (probable pneumonia), PVAP and undetermined
infections was 12.9, 7.1, 3.7and 2.1 per 1000 ventilator days respectively. The study did not assess
the agreement of surveillance tools, but the authors mention that 41 patients met the PrVAP and
PVAP tier of the new VAE surveillance versus nine patients using PNU/VAP.
A study by Taylor et al. (2014) evaluating two surveillance tools in 285 patients in a single PICU
revealed that seventeen patients met the PVAP tier using the new VAE surveillance tool, and 15 met
the VAP criteria using the older PNU2/VAP surveillance tool. However, only nine patients were
detected by both tools. The comparison between proportions of patients who met the PVAP tier by
the new VAE surveillance tool and those who met the VAP by PNU/VAP surveillance tool was not
significant (p= 0.78).
The prospective study by Narayanan et al. (2016) demonstrated that the new VAE surveillance tool
was unable to identify any additional VAP cases to those detected using the PNU/VAP surveillance
tool. In this study, children (n=325) were prospectively evaluated over a six-month period and it was
found that seven met the VAC tier. Out of these seven children, six met the IVAC tier and from those
six, three met the PVAP tier. The incidence rate for VAC and PVAP was 4.2 and 1.8 per 1000
30
ventilator days respectively and the VAP rate using the PNU/VAP surveillance tool was 2.4 per 1000
ventilator days.
Three studies evaluated the new VAE surveillance tool but with modification of some criteria or
parameters for a paediatric sample (Beardsley et al., 2016; Cirulis et al., 2016; Cocoros et al., 2016).
A multi-centre retrospective cohort study undertaken by Cocoros et al. (2016) involved 8862 patients
across five hospitals. To detect VAC, instead of using a daily minimum PEEP value increase of at
least 3cm H2O in the adult VAE surveillance tool, they replaced the PEEP value with Mean Airway
Pressure (MAP) 4, 5, 6, and 7 cmH2O. Furthermore, they tested an increment of daily minimum FiO2
into three thresholds, 0.20, 0.25 and 0.30, instead of only the daily minimum FiO2 0.20 threshold used
in the adult VAE surveillance tool. Using the MAP≥ 4/FiO2 ≥ 0.20, the VAC incidence rates were
3.3 to 4.6 per 1000 ventilator days. The cardiac ICU patients had a higher incidence rate as compared
to PICU and Neonatal ICU. Using the MAP ≥4/ FiO2 ≥ 0.20, the incidence rates were 2.9 to 3.2 per
1000 ventilator days. Using MAP ≥7/ FiO2 ≥ 0.30, the incidence rates were 1.1 to 1.3 per 1000
ventilator days (Cocoros et al., 2016). The main findings of the study supported the position that the
tool was able to detect VAC regardless of thresholds and was associated with higher morbidity and
mortality with hazard ratios ranging from 1.6 (95% CI, 0.7-3.4) to 6.8 (2.9-16.0), depending on ICU
types (four PICUs, three cardiac ICUs and one Neonatal ICU). The study proposed the application of
FiO2 of 0.25 and MAP of 4 cmH2O thresholds to identify VAC in paediatric patients (Cocoros et al.,
2016).
Beardsley et al. (2016) applied a daily minimum PEEP of at least 2cmH2O instead of 3cmH2O used
in the adult VAE surveillance tool to evaluate VAC in 300 episodes of mechanical ventilation (217
PICU patients). They also evaluated various ventilator-associated infections criteria such as
PNU/VAP (Horan et al., 2008), ventilator associated tracheobronchitis (VAT) (Craven, Chroneou,
Zias, & Hjalmarson, 2009; Tamma et al., 2011) and lower respiratory tract infection criteria (LRTI)
(Horan et al., 2008). The results showed that the VAE surveillance tool used was consistent with
PNU1/VAP surveillance tool. The incidence of IVAC was 2.16 per 1000 ventilator days (four
mechanical ventilation episodes) that met the VAE surveillance tool versus the incidence of VAP,
which was 2.60 per 1000 ventilator days (five mechanical ventilation episodes) that met the
PNU/VAP surveillance tool. However, only one mechanical ventilation episode met both surveillance
tools. The incidence of VAT and LRTI was 5.19 (four mechanical ventilation episodes) and 6.92 (16
mechanical ventilation episodes) per 1000 ventilator days respectively.
31
Cirulis et al. (2016) found high levels of specificity of both VAE surveillance tools — the new VAE
surveillance tool described by the authors as VAC0 and a modified VAE surveillance tool as VACMP
— in paediatric traumatic brain injury patients. The VACMP used a modification of VAC/VAE criteria
for a PEEP value greater than or equal to 2cmH2O, sustained for more than or equal to one day, and
retained other parameters for IVAC, and PVAP (VAC0) defined in the same way as the new VAE
surveillance tool. Previously they assessed 119 children using the PNU2/VAP surveillance tool and
reported that 39 patients met the VAP criteria. Nine patients met the new VAE surveillance tool
(VAC0) tier and 22 patients met the modified VAE surveillance tool (VACMP). Despite high
specificity of both VAE surveillance tools (VAC0 and VACMP), low sensitivity was demonstrated in
comparison to PNU2/VAP surveillance tools. Furthermore, patients who experienced pulmonary
diagnosis in VAE or VAP using PNU2/VAP tools had significantly worse outcomes compared to the
group who were without respiratory complications.
What are the potential challenges in the application of the VAE surveillance tool in paediatric
clinical practise?
The principal challenge in the application of the VAE surveillance tool in the paediatric population
relates to unique paediatric physiology and the resulting differences in ventilation modalities
clinically required (Cocoros et al., 2016). The new VAE surveillance tool excludes all patients on
high-frequency oscillatory ventilation (HFOV) or extracorporeal life support (Centers for Disease
Control and Prevention (CDC) & National Healthcare Safety Network (NHSN), 2015b).
Furthermore, during periods of time while the patient is on airway pressure release ventilation, the
VAC should be determined by the changes in FiO2 only (Centers for Disease Control and Prevention
(CDC) & National Healthcare Safety Network (NHSN), 2015b).
One of the benefits highlighted in the adult literature is that the tool of VAE enables automated data
extraction which reduces time spent on surveillance and minimises human bias (Klein Klouwenberg
et al., 2014). Only Phongjitsiri et al. (2015) reported a locally developed and supported system that
enabled automatic data extraction from electronic medical records for later analysis and which had
minimal workforce implications. Beardsley et al. (2016) identified mechanical episodes in eligible
patients using the Virtual PICU Systems from electronic medical records; however, the data were
limited to demographic characteristics, Paediatric Risk of Mortality, and PICU outcomes, such as
duration of mechanical ventilation, PICU length of stay, and PICU mortality. While Beardsley et al.
(2016) briefly explained the data retrieval for their study, other researchers used data from electronic
medical records without automated data retrieval (Cirulis et al., 2016; Cocoros et al., 2016; Iosifidis
32
et al., 2016; Narayanan et al., 2016; Taylor et al., 2014). Thus, challenges remain regarding the
feasibility and reliability of both automated and manual collection and documentation of clinical data
(Magill et al., 2014).
DISCUSSION
Surveillance tools for assessing VAP and other associated adverse outcomes have existed since 1988.
Each new iteration builds on previous models, yet none have been designed exclusively for
paediatrics. This review explored the literature comparing the current VAE tool with previous
iterations in paediatrics. Only one study undertook sensitivity and specificity testing between the two
tools; it found that although there was high specificity, the low sensitivity score indicated that
underreporting of VAE may occur with the new surveillance tool (Cirulis et al., 2016). Several other
studies compared calculated incidence rates across two tools, and these authors concluded that there
was little agreement between the two tools, although there was consistency in reporting (Beardsley
et al., 2016; Iosifidis et al., 2016; Narayanan et al., 2016; Taylor et al., 2014). This has significant
ramifications for paediatric settings as the number of VAE cases meeting the VAE surveillance tool
in children seems to be higher than those in adults. Of 1209 adult medical and surgical patients, 67
met the VAC tier, and of those, 34 met the IVAC tier with an incidence rate of 7.0 and 3.6 per 1000
ventilator days respectively (Boyer et al., 2015). In contrast, the PICU study by Phongjitsiri et al.
(2015), reported an incidence rate of 20.9 and 12.9 episodes per 1000 ventilator days respectively.
Surveillance tools act not only as a means of determining the quality of care delivered and measuring
patient outcomes but also as a benchmark to compare outcomes across units and countries (Haustein
et al., 2011; Klompas, 2013a).
Innovation in care delivery and effectiveness in preventative strategies can be measured based on
surveillance data, but if the data are uncertain, we should question if they are adequately robust to
draw conclusions. Substantial variation exists in incidence rates, with more VAE identified in a study
conducted by Iosifidis et al. (2016), (11.2 episodes per 1000 ventilator days), compared with a
paediatric study that was published in the same year by Cocoros et al. (2016), (2.9 to 3.2 episodes per
1000 ventilator days). Variation in case mix provides a possible explanation for this range,
particularly with a broader group of patients (infant and paediatric) and inclusion of all types of ICUs
by Cocoros et al. (2016). Adult multicentre ICU studies report similar outcomes (Bouadma et al.,
2015; Stevens et al., 2014). Consistent results were reported in the three included studies which
adopted the new VAE tool (Narayanan et al., 2016; Phongjitsiri et al., 2015; Taylor et al., 2014),
compared with two studies where modifications were made to the criteria of the VAE surveillance
tool (Beardsley et al., 2016; Cirulis et al., 2016).
33
The inclusion of positive microbiological cultures of endotracheal aspirate in PNU1/VAP by some
researchers is questionable, as these may not reflect true infection (Chang & Schibler, 2016). There
was consistency regarding the number of PVAP-VAP cases meeting the criteria, 12 versus 13
children using the VAE surveillance tool and the PNU/VAP surveillance tool, respectively; however,
the VAE surveillance tool failed to identify eight patients with positive tracheal cultures (Iosifidis et
al., 2016). A similar result was demonstrated in a comparison study of the traditional PNU/VAP
versus VAE surveillance tool in adult patients. The VAE surveillance tool revealed the incidence of
10.0 episodes per 1000 ventilator days versus 8.0 episodes per 1000 ventilator days using the
traditional PNU/VAP surveillance tool; however, the VAE surveillance tool discovered 32% more
patients with VAP because the tool signalled other causes such as fluid overload (Klein Klouwenberg
et al., 2014). The historical PNU/VAP surveillance tools relied heavily on radiological findings and
the subjective interpretation of respiratory signs and symptoms. However, the new VAE surveillance
tool replaces these criteria with objective parameter tools such as FiO2 and PEEP, thus minimising
subjectivity and potentially reducing ambiguity in the diagnosis of VAP and improving both internal
and external validity (Klompas, 2010, 2012).
Another challenge unique to paediatric critical care clinicians is using adult-focused tools and
mechanisms to assess the prevalence of iatrogenic adverse events. The review of the current evidence
illustrates that this is the case with diagnosing VAEs. This suggests the merit of the new VAE
surveillance tool, which has broader capture of VAE in the paediatric population and opportunities
to examine clinical impact and later, the specific preventative strategies (Muscedere et al., 2013;
Septimus et al., 2015). Both tools have merit in identification of ventilator-associated complications,
but the new VAE surveillance provides other explanations for non-infectious complications that may
exist apart from VAP in mechanically ventilated patients. A recent paediatric study examined the
traits of VAE, finding 44% were also due to non-infectious conditions such as atelectasis and
pulmonary oedema and shock (Phongjitsiri et al., 2015) (compared with pneumonia, pulmonary
oedema, and acute respiratory distress syndrome in adults) (Boyer et al., 2015; Hayashi et al., 2013;
Klein Klouwenberg et al., 2014; Klein Klouwenberg et al., 2013).
The inclusion and exclusion of modality of ventilation and antibiotic treatment for VAE surveillance
tools in paediatrics appears somewhat uninformed. Perhaps, in the paediatric context, the difference
in ventilator treatment modalities and variation in antibiotic prescribing may not be as profound as
that which is evident in adult settings (Beardsley et al., 2016; Cocoros et al., 2016). However, omitting
HFOV risks excluding very sick children as this modality is one of the recommendations in children
with acute respiratory failure (Santschi, Randolph, Rimensberger, & Jouvet, 2013). Cocoros et al.
34
(2016), argued for the need to include patients on HFOV, considering that the usage of HFOV in
paediatric patients is more frequent than in adult units, while other studies did not specify their
ventilation inclusion and exclusion criteria (Beardsley et al., 2016; Cirulis et al., 2016; Iosifidis et al.,
2016; Narayanan et al., 2016; Taylor et al., 2014). The modification threshold of PEEP to 2 cmH2O
was also applied, and an adjustment period of stability to 24h was also tested (Beardsley et al., 2016;
Cirulis et al., 2016). However, the evidence of these modifications is limited to three studies.
The value of automated data extraction using the adult VAE surveillance tool is also worthy of further
consideration in paediatric settings. In an adult study, the automated data extraction was shown to be
not only efficient but also to increase reliability and objectivity (Stevens et al., 2014). The potential
benefit of this is also acknowledged in paediatric studies (Phongjitsiri et al., 2015; Taylor et al., 2014).
A recent adult study confirmed that automated data extraction is feasible with 100% sensitivity and
accuracy when compared with the manual method (Hebert, Flaherty, Smyer, Ding, & Mangino,
2017). Collaboration between clinicians and experts in medical information and systems technology
may result in an innovative data extraction platform (Magill et al., 2013).
IMPLICATIONS FOR FUTURE RESEARCH
The studies identified in this review are largely from the United States where VAP and VAE
surveillance is a key clinical performance indicator. Thus, it may not be surprising that to date, there
is limited research related to the application of the new VAE surveillance in paediatrics in other
countries. However, discrepant VAP rates reported across the globe in adults and children may be
due in part to a lack of objectivity in the existing VAP surveillance tool, (Klompas, 2010) and further
research in investigating VAE in paediatrics is warranted. Although there is promising evidence of
the benefits of using the new adult VAE surveillance tool (Magill et al., 2014; Muscedere et al., 2013),
there is an urgent need to conduct more studies to achieve a paediatric version of the VAE surveillance
tool.
LIMITATIONS
The review was restricted to English language literature, and grey literature was not included in the
search strategy. This may have introduced selection and publication biases.
CONCLUSION
This is the first review of the application of the VAE surveillance tool compared with historical
surveillance data in paediatrics. The strength of evidence is currently of a low level. There is
substantial variation cited across studies and a lack of agreement between the old and new tools when
35
applied to clinical data. This suggests that the current paediatric surveillance tool does not fully
capture the prevalence of VAE in paediatrics, although it is adequate for monitoring. We caution
against comparing the current monitoring results with historical VAP data. Although limited to only
seven studies, this review provides insights into published literature related to the application of the
new VAE surveillance tool and the potential implementation challenges in the paediatric population.
SUMMARY
This chapter shows that the VAE surveillance tool for paediatric patients is currently being developed.
Unlike in adult centres, using this tool is not mandated. A possible reason for this is that the
surveillance tool needs further refinement to suit the paediatric population (Cocoros et al., 2016). It
is also necessary to consider the complexities of the pathophysiology and variations of the underlying
problems in children who require additional attention before the decision on whether to adapt and/or
adopt the criteria currently used in adults can be made (Aelami et al., 2014; Cocoros, Priebe, Gray,
et al., 2017; Cocoros, Priebe, Logan, et al., 2017).
Published data pertaining to adults demonstrated that the VAE surveillance tool showed promising
results in the ability of the surveillance tool to identify other ventilator-associated complications,
improving internal and external validity and enabling/facilitating automated clinical data extraction
(Boyer et al., 2015; Klein Klouwenberg et al., 2014; Klein Klouwenberg et al., 2013; Magill et al.,
2013; Stevens et al., 2014). Despite being accepted in the case of adult ventilated patients, limited
information is known about the applicability of the VAE surveillance tool in children (Lutmer &
Brilli, 2016) and more studies are warranted. Currently, there is no available information regarding
VAE for paediatric patients in the Australian context.
Summary
This chapter has demonstrated that data describing the epidemiology of VAP and associated risk
factors vary globally. The impact of VAP is considerable and challenges healthcare providers. The
lack of paediatric epidemiological data for VAE is evident, particularly in the Australian context.
Little is known about the pathogenesis of VAE or the risk factors for VAP, in particular the modifiable
risk factors. There is limited evidence of the application of the new VAE surveillance tool in children.
Only one study has attempted to evaluate the sensitivity and the specificity of VAE in a paediatric
setting. Thus, more research is required to verify the applicability of the VAE tool on the paediatric
population.
36
The next chapter describes the review of literature relating to preventative strategies, compliance
issues and VAP education.
37
Literature Review: Preventative strategies, compliance and VAP education
Introduction
This chapter will present the review of literature for preventative strategies (VAP and VAE) in
paediatric patients and the challenges with VAP prevention compliance for healthcare workers. This
chapter also includes current information for VAP education and the role of parents and nursing staff,
particularly concerning ‘Speaking up for hand hygiene’.
Preventative strategies for VAP
VAP preventative strategies, also known as ‘VAP bundles’, are a group of evidence-based
interventions designed to create better patient outcomes and are implemented as a set of interventions
rather than on an individual intervention basis (Resar et al., 2014; Resar et al., 2005). This concept
was introduced by the Institute for Healthcare Improvement (IHI) in 2002 in the USA and was
followed in 2004 by the “The 100,000 Lives Campaign” which included ventilator bundles in adult
ICUs. This bundle consisted of four preventative strategies: (1) head of the bed (HOB) elevation 30º-
45º; (2) peptic ulcer disease (PUD) or gastrointestinal (GI) prophylaxis; (3) daily sedation vacation
and assessment of readiness to extubate; and (4) deep vein thrombosis (DVT) prophylaxis (American
Thoracic & Infectious Diseases Society, 2005; Gandra & Ellison, 2014). Extensive research has been
carried out in the adult population on the effectiveness of the VAP bundle. The evidence has
supported the use of the VAP bundle with modifications over time. For example, oral care with
chlorhexidine 0.12% was added in 2010 (Munro & Ruggiero, 2014).
Research into the effectiveness of VAP bundles in paediatrics was first published in 2007 (Cooper &
Haut, 2013). The IHI outlined a supplement for VAP bundles in children in 2005. This consists of
four preventative strategies which are a modification of the adult VAP bundle: (1) HOB to 15º-30º for
neonates, 30º-45º for infants or above; (2) daily assessment of readiness to extubate (daily sedation
interruption is not recommended due to high risk of unplanned extubation); (3) PUD or GI
prophylaxis (only as appropriate for the age and condition of the child); and (4) deep vein thrombosis
(DVT) prophylaxis (unless contraindicated, and as appropriate for the age and condition of the child)
(Resar et al., 2014). In addition to these four preventative strategies, three other preventative strategies
were also proposed, as detailed in Table 3.1.
38
Table 3.1: Additional preventative strategies to consider for the paediatric VAP bundle
No Preventative strategy Details
1 Oral care 1. Daily comprehensive oral care according to age and
condition of the child
2. Increase the frequency of oral care to two-hourly for
‘high risk’ patients
3. Include the use chlorhexidine for children older than two
months of age
2. Ventilator circuit checks 1. Heated ventilator circuit should be used to reduce the
amount of condensate
2. Accumulated water in the ventilator circuit should be
drained every 2-4 hourly and prior to patients being re-
positioned
3. Ventilator circuits should be changed when visibly
soiled
4. Hand hygiene performed before and after contact with
ventilator circuits
3. Aspiration devices 1. Oral aspiration device (example Y-suction catheter)
should be kept in a clean non-sealed plastic bag when not
in use
2. Single-use suction catheter
3. All suction related equipment should be changed when
it is soiled or otherwise indicated
Even though the IHI has outlined the supplement for the ventilator bundle in paediatric patients, the
evidence is sparse (Cooper & Haut, 2013). Variations exist between hospitals, and adult preventative
strategies are often adopted into paediatric settings without further testing (Coffin et al., 2008; Curley
et al., 2006). Table 3.2 presents the studies undertaken in children which utilise variations of
individual VAP preventative strategies (in a bundle) in their study settings and the subsequent VAP
rates.
39
Table 3.2: Individual VAP preventative strategies used in paediatrics and VAP rates (in order of publication date)
Authors Country
number of
patients
Study design HOB Daily
assess for
extubate
GI/
DVT
prophylaxis
Other strategies
Pre/baseline
to Post VAP
rates*
Brilli et al.
(2008)
USA n =26 Retrospective
matched case-
control study
/ 1. Daily oral care with chlorhexidine 7.8 to 0.5a
Bigham et
al. (2009)
USA n =1782 Pre-post
implementation
design
(30- 450)
/ 1. Hand hygiene before and after contact with
ventilator circuit
2. Oral care every 2-4 hourly
3. Drain ventilator circuit 2-4 hourly and before
repositioning patient (use heated wire circuits to
reduce rainout)
4. Use endotracheal tube with subglottic suction
for children more than 12 years old
5. Change ventilator circuits and in-line suction
catheters only if visibly soiled
7. Store oral suction devices in non-sealed plastic
bag at the bedside when not in use, (rinse devices
after use)
8. Wear a gown before providing care to patients
when soiling from respiratory secretions is
expected
5.6 to 0.3
Brierley et
al. (2012)
UK n= 730 Pre-post
implementation
design (quality
improvement)
(20-300)
(Ranitidine
for those
patients not
on full
feeds/
1. Oral care with antiseptic 4-hourly or 12-hourly
with a toothbrush
2. Clean suctioning equipment daily or once
indicated
3. Routine chest x-ray interpreted by
physiotherapist only
4. Documentation of individual implemented
VAP preventative strategies to be completed
every 4-hours
5.6 to 0.0
40
5. Documentation of indication of VAP
compliance needs to be completed every shift
Rosenthal,
Alvarez-
Moreno,
et al.
(2012)
Colombia,
Philippines,
India, El
Salvador,
Turkey n=
4339
Pre-post
implementation
design
(30- 450)
/ 1. Hand hygiene
2. Minimising duration of mechanical ventilation
by use of non-invasive ventilation
3. Preference of orotracheal tube over the
nasotracheal tube
4. Maintenance of ETT cuff pressure of at least
20cm H2O
5. Removal of condensate from ventilator circuits,
keeping the ventilator circuit closed during
condensate removal
6. Changing of ventilator circuit only when
visibly soiled
7. Avoid gastric over-distension
8. Avoidance of histamine-receptor 2 –blocking
and proton pump inhibitor
9. Use sterile water to rinse reusable respiratory
equipment
11.7 to 8.1
Obeid,
Naous,
Naja, and
Naja
(2014)
Lebanon n=
107
Pre-post-
implementation
design
(15- 300)
/ 1. Hand hygiene
2. Oral hygiene with antiseptic solution
3. Oro-gastric residual volume measurement
before feeding
4. Closed suction system
52.6 to 6
cases/100
patients
De
Cristofano
et al.
(2016)
Argentina
n=713
Quasi-
experimental
uninterrupted
time series
(every six
months for 2
years)
(> 300)
/ 1. Oral hygiene with chlorhexidine 0.12% six-
eight hourly
2. Daily sedation interruption
3. Clean and dry ventilator circuits
6.3, 5.7, 3.2,
1.8 to 0.0
*= /1000 ventilator days) (pre/baseline to post); a=control, case
41
Evidence of individual preventative strategies within the VAP bundle
Variations in VAP bundles exist between institutions (Cooper & Haut, 2013; Shmilev & Yankov,
2012). The following section further explains data for individual elements that are commonly
included in ventilator/VAP bundle.
Hand Hygiene
Hand hygiene is the least costly and most effective measure in the prevention of HAIs across
healthcare settings (Kozlowski, 2012; World Health Organization (WHO), 2009b). With the recent
spread of multi-resistant organisms in hospitals, hand hygiene has become significant for researchers
and healthcare workers alike (Rabelo et al., 2014). This is because poor hand hygiene may initiate
bacterial cross-contamination, contributing to HAIs and VAP (Craven, 2006; World Health
Organization (WHO), 2009a). Study have shown that common VAP causative agents such as
Staphylococcus aureus and other gram-negative bacilli organisms cross transmitted to patients
through hand contact of healthcare workers (Sachdev et al., 2013; Safdar Crnich & Maki, 2005).
In 2005, the WHO initiated a campaign entitled, “Save Lives: Clean Your Hands” which outlined the
five key instances of hand hygiene for healthcare workers. This approach recommends healthcare
workers clean their hands in the following instances:
(1) before touching a patient;
(2) before cleaning/aseptic procedures;
(3) after body fluid exposure;
(4) after touching a patient; and
(5) after touching patient surroundings (World Health Organization (WHO), 2009b).
From the perspective of infection control, hand hygiene is considered as general prevention (Gandra
& Ellison, 2014) and often not included in adult ventilator bundles (Resar et al., 2014). Some studies
have recommended incorporation of hand hygiene into the ventilator bundle (Bouadma, Mourvillier,
Deiler, Le Corre, et al., 2010; Tolentino-DelosReyes, Ruppert, & Shiao, 2007). In paediatric and
neonatal settings, hand hygiene becomes instrumental in the prevention of infection (Azab et al.,
2015; Obeid et al., 2014), as children’s immature host defence makes them more susceptible to
infection compared to adults (Chhapola & Brar, 2015). For instance, two paediatric pre-post
implementation study designs incorporating hand hygiene into their ventilator bundle by Bigham et
al., (2009) and multicentred study by Rosenthal, Alvarez-Moreno, et al. (2012) revealed a reduction
of VAP incidence from 5.6 to 0.3 and 11.7 to 8.1 per 1000 ventilator day. More recent findings Obeid
et al., (2014) reported reduction from 53 cases to six cases/100 patients.
42
In PICU, hand hygiene is identified as a key target for quality improvement efforts (Harris et al.,
2011). Some studies have shown that hand hygiene supports the reduction of VAP rates. For example,
Koff, Corwin, Beach, Surgenor, and Loftus (2011) showed that the new hand hygiene initiative
directly resulted in a reduction of VAP from 6.9 to 3.7 per 1000 ventilator days. Similarly, studies in
paediatrics and neonatal settings (with hand hygiene as one of their preventative strategies) showed
the VAP rate significantly decreased from 11.7 to 0.3 per 1000 ventilator days (Azab et al., 2015;
Bigham et al., 2009; Rosenthal, Alvarez-Moreno, et al., 2012). Some studies have shown that the
VAP rate reduced to 0 per 1000 ventilator days (Brierley et al., 2012; De Cristofano et al., 2016).
Therefore, hand hygiene is the most important element in paediatric VAP preventative strategies
Oral Hygiene
Poor oral hygiene is associated with bacterial colonisation and dental plaque formation in the oral
cavity and may place patients at higher risk of VAP (Fourrier et al., 2005; Klompas, Branson, et al.,
2014; Shmilev & Yankov, 2012; Tablan et al., 2004). Inadequate oral hygiene leads to bacterial
plaque formation and oral biofilms that potentially contribute to VAP (Munro et al., 2006; Paju &
Scannapieco, 2007).
A randomised controlled trial by Berry, Davidson, Masters, Rolls, and Ollerton (2011) in adult ICUs
compared two oral care strategies to test the effects of three regimes on dental plaque colonisation
and the incidence of VAP (group A: two-hourly oral rinse with sterile water (control); group B:
sodium bicarbonate mouth wash two-hourly, and Group C: twice daily irrigations with
chlorohexidine 0.2% aqueous oral rinse and two-hourly irrigations with sterile water). All regimes
involved cleaning with a toothbrush and non-foaming toothpaste. The study suggested that
standardised oral hygiene protocols may improve outcomes in critically ill patients although it did
not establish a significant difference between groups (Berry, Davidson, Masters, et al., 2011).
Oral hygiene requires age appropriate standardisation to ensure maximum benefit. This includes oral
assessments prior to oral care to determine the frequency of oral hygiene performance and the type
of cleaning solution used (Oral Care Guideline/Protocol) (Bigham et al., 2009; Brierley et al., 2012;
Johnstone, Spence, & Koziol-McClain, 2010). A critical appraisal of 14 relevant literature sources of
oral hygiene in children by Johnstone et al. (2010) produced PICU oral care guidelines. The
guidelines are shown in Table 3.3.
43
Table 3.3: Oral hygiene protocol for mechanically ventilated children
Age categories Oral hygiene protocol; frequency of performance
Neonates and infants with no
teeth
(a) Moisten mouth with swabs soaked in physiological saline; 2-hourly
(b) Coat lips with petroleum jelly; 2-hourly
Infants and children < 6 years
with teeth
(a) Brush teeth with small, soft toothbrush and fluoride toothpaste; suction out
excess toothpaste, but do not rinse out mouth; 12-hourly
(b) Moisten mouth with swabs soaked in physiological saline; 2-hourly
(c) Coat lips with petroleum jelly; 2-hourly and as needed
Children ≥ 6 years with teeth (a) Brush teeth with small, soft toothbrush and fluoride toothpaste; suction out
excess toothpaste, but do not rinse out mouth; 12-hourly
(b) Rinse mouth with 0.1% chlorhexidine: irrigate with a syringe or wipe oral
mucosa with a swab; suction excess solution, but do not rinse out mouth with
water; use at least 30 minutes after brushing teeth; 12-hourly
(c) Moisten mouth with swabs soaked in clean water or physiological saline;
2-hourly
(d) Coat lips with petroleum jelly; 2-hourly
A standardised oral hygiene protocol inclusive of oral assessment is strongly recommended. A
systematic review by Gibson et al. (2010) evaluating oral assessment tools found that the most
effective tool in paediatric intensive care is the Oral Assessment Guide (OAG) (Eilers, Berger, &
Petersen, 1988). This tool detects changes in oral status for clinical assessment and guides nursing
interventions, in particular to the selection of solutions to be used for oral hygiene. Evidence supports
the use of a soft, small toothbrush rather than foam swabs to remove oral plaque (Munro et al., 2006;
Pearson & Hutton, 2002) and advises against the use of tap water for mouth rinsing due to the
potential presence of pathogenic bacteria (Berry, Davidson, Masters, & Rolls, 2007). An example of
an oral hygiene protocol introduced in a paediatric study by Brierley et al. (2012) consisted of an
initial oral assessment and four-hourly oral care using sterile water, chlorhexidine, and twelve-hourly
toothbrush and toothpaste, based on patient age. In this study, the VAP rate reduced from 5.6 to 0.0
per 1000 ventilator days after the VAP bundle implementation.
Use of chlorhexidine 0.12% for oral care for the prevention of VAP has been previously advocated
in the adult literature (Fourrier et al., 2005), but a recent adult study by Klompas, Li, Kleinman,
Szumita, and Massaro (2016) contradicted this finding. In this observational study, the researchers
found that oral hygiene with 0.12% chlorhexidine was associated with an increased hazard of
ventilator mortality; hazard ratio (HR), 1.63; 95% CI, 1.15-2.31; p= 0.006.
44
In paediatrics, two randomised controlled trials found that the use of chlorhexidine 0.12% for oral
care did not reduce the incidence of VAP (Jácomo, Carmona, Matsuno, Manso, & Carlotti, 2011;
Kusahara, Peterlini, & Pedreira, 2012). In a prospective, randomised, double-blind, placebo-
controlled trial of 160 children undergoing surgery for congenital heart disease, participants were
randomised into two groups: experimental group n=87 received 0.12% chlorhexidine, and the control
group n=73 received a placebo (Jácomo et al., 2011). Each child received their respective oral care
solution preoperatively and twice daily after surgery until PICU discharge or death. No significant
difference in VAP rates was observed (p=0.57). A prospective randomised controlled and double
blinded study by Kusahara et al. (2012), was undertaken among 96 children. The intervention group
(n=46) received chlorhexidine 0.12% with a toothbrush and an antiseptic gel twice daily. The placebo
group (n=50) received oral care with a toothbrush and non-antiseptic gel twice daily. Both groups
were comparable to each other with no difference in VAP rates (p=0.95).
The concentration of chlorhexidine potentially needs to be tailored to the critically ill child’s need;
however, to date no randomised controlled trial has been carried out to test this hypothesis. In
summary, recent evidence shows that the use of 0.12% chlorhexidine does not seem to impact VAP
rates in children (Jácomo et al., 2011; Kusahara et al. (2012).
Endotracheal Suctioning
The purpose of endotracheal suctioning is to maintain airway patency by removal of secretions
(Morrow & Argent, 2008; Volsko, 2013). Endotracheal suctioning with the assistance of
humidification of inspired gas in PICU patients remains the main intervention for secretion
management (Evans, Syddall, Butt, & Kinney, 2014). A review of the indications for endotracheal
suctioning in paediatrics recommends endotracheal suctioning to improve respiratory mechanics and
that it should be performed as indicated by the presence of secretions (Morrow & Argent, 2008). In
a qualitative study, Davies, Monterosso, and Leslie (2011), identified five top indications for
performance of endotracheal suctioning, namely; (1) suspected ETT obstruction due to secretions,
(2) audible or visible secretions, (3) decreased oxygen saturations, (4) suspected aspiration and (5)
dyspnoea or signs of respiratory distress.
When the endotracheal suctioning procedure is required, attention must also be given to the use of
aseptic techniques to prevent contamination while handling suctioning equipment (Tolentino-
DelosReyes et al., 2007). One study provided supportive evidence for reuse of suction catheters in
children. A randomised controlled trial compared the incidence of VAP in patients who were
suctioned using the same catheter in all suctioning episodes within 24 hours (intervention n=241) and
45
patients who were suctioned using a new catheter for each suction episode (control group n=245)
(Scoble, Copnell, Taylor, Kinney, & Shann, 2001). No difference of VAP rates were found; 14
patients developed VAP in the intervention group and 12 patients in the control group. This study
supported reusing the catheter for 24 hours as a cost saving measure.
A comprehensive review of literature by Tume and Copnell (2015) concluded that limited evidence
exists regarding suctioning practise in paediatrics, especially through adequately powered studies to
compare closed and open suction methods in children and VAP occurrence. A prospective
observational study in ventilated infants and children by Morrow, Mowzer, Pitcher, and Argent
(2012) assessed the frequency of VAP on closed versus open systems and was undertaken with 263
PICU admissions. The frequency of VAP for patients on closed versus open systems was 20.3% and
23.3% respectively, with no significant difference between the two methods (p=0.60). Studies in
adults reached similar conclusions; no significant difference was found between the two methods
(Lorente et al., 2005; Topeli, Harmanci, Cetinkaya, Akdeniz, & Unal, 2004).
In terms of the effectiveness of secretion removal using the two suction systems, a study conducted
using an animal model found that the closed system was less effective compared to open system
regardless of ventilator mode (conventional or high frequency ventilation) (Copnell et al., 2007). In
addition, the nature of a closed suction system does not require circuit disconnection from the patient
that may allow for condensate water/fluid from the circuit to accumulate, and this may promote
bacterial colonisation compared to an open suctioning system (Shmilev & Yankov, 2012).
A prospective study by Owen et al. (2016) examined adverse events associated with saline instillation
during ETT suctioning. The results demonstrated hemodynamic instability, bronchospasm and
oxygen desaturation were the adverse events most reported (p<0.001) and three of the 62 patients
were diagnosed with VAP (4.8 cases/100 mechanically ventilated patients). The study suggested
cautious use of saline during ETT suctioning in paediatric patients. An integrative review by Schults,
Mitchell, Cooke, and Schibler (2018), concluded that the instillation of saline during ETT suctioning
was not associated with the development of VAP in children.
Available evidence demonstrates no significant difference between open suctioning and closed
suctioning systems in regards to VAP occurence. To date no specific recommendations from the CDC
have been made on the closed suctioning system (Klompas, Branson, et al., 2014). According to
Tume and Copnell (2015), a closed suction system should be used after an individual patient
46
assessment is performed. Regardless, the maintaining of aseptic technique while performing ETT
suctioning is still advocated as a VAP preventative strategy.
Head of the bed (HOB) Elevation
Systematic reviews in adult patients are inconclusive in their support for the clinical benefits of HOB
elevation and the importance of the degree of elevation, despite this being an IHI recommended
inclusion in the ventilator bundle (Niel-Weise et al., 2011; Resar et al., 2014). Nevertheless, the
practise of elevating the HOB to 30-45 degrees has been a standard practise in nursing for many
years, except when clinically contraindicated (Munro & Ruggiero, 2014; Tablan et al., 2004; Wip &
Napolitano, 2009). In the systematic review by Niel-Weise et al. (2011), the recommendation for
HOB elevation was 20-45 degrees. This is supported by earlier observational and randomised
controlled trials which demonstrated that the supine position during 24 hours of mechanical
ventilation was an independent risk factor for VAP (adjusted OR=2.9; 95% CI, 1.3 to 6.8; p=0.013)
(Cook & Kollef, 1998) and that HOB elevation was associated with a reduction in the risk of
aspiration (Drakulovic et al., 1999).
Recommendations for the degree of HOB elevation are more complicated in infants and children.
This is due to the significant challenge in maintaining the correct position in small children,
particularly if lightly sedated (Foglia et al., 2007). Paediatric studies suggest that the HOB should be
kept at a minimum of 15-45 degrees (Bigham et al., 2009; Brierley et al., 2012; De Cristofano et al.,
2016; Obeid et al., 2014; Rosenthal, Alvarez-Moreno, et al., 2012).
Despite the low level of evidence supporting the rationale and degree of HOB elevation in paediatric
ventilated patients, this procedure is considered clinically important (Foglia et al., 2007; Juneja et al.,
2011). HOB elevation also offers benefits by minimising the risk of atelectasis by improving lung
expansion (Niel-Weise et al., 2011) and helps to reduce the severity of gastrointestinal reflux, hence
minimising the possibility of micro-aspirations that may lead to VAP (Shmilev & Yankov, 2012).
Endotracheal tube (ETT) and cuff pressure checks
The intubation procedure can compromise the natural barrier between the oropharynx and the trachea
and may promote colonisation and the migration of pathogenic organisms into the lower respiratory
tract (Berry, Davidson, Nicholson, Pasqualotto, & Rolls, 2011; Scannapieco et al., 2009). Intubation
with cuffed endotracheal tubes has become routine practise over the last two decades in paediatric
patients less than eight years of age (Eschertzhuber et al., 2010). Advancements in cuff technology
(microcuffs) have improved the safety of this type of ETT in infants and young children
47
(Eschertzhuber et al., 2010; Tobias, Schwartz, Rice, Jatana, & Kang, 2012; Weiss & Dullenkopf,
2007). Air leakage above the microcuff is an important factor in minimising potential microaspiration
of pathogenic secretion into the lungs (Dave, Frotzler, Madjdpour, Koepfer, & Weiss, 2011). Hence,
an optimal seal at minimal pressure is required to protect the airway and minimise VAP potential.
A literature review by Bhardwaj (2013) concluded that an ETT with a cuff pressure of ≤15cm H2O
in children ensures an adequate seal, while Dullenkopf, Bernet‐Buettiker, Maino, and Weiss (2006)
suggested keeping a maximum pressure of 20cm H2O. Rosenthal, Alvarez-Moreno, et al. (2012) in
their study involving five developing countries including Colombia, Philippines, India, El Salvador,
and Turkey suggested keeping ETT cuff pressure to a minimum of 20cm H2O in children.
Weiss et al. (2009) studied 2,246 children (aged from birth to five years old) with cuffed and uncuffed
ETTs in a prospective randomised controlled trial. They found that the use of cuffed ETTs in children
provided a reliably sealed airway at cuff pressures of ≤ 20cm H2O, reducing accidental extubations
and minimising the risk of post-extubation stridor. The study also suggested that the minimum cuff
pressure required to seal the trachea was 10.6cm H2O (Weiss et al., 2009).
No specific study has evaluated the frequency of ETT cuff monitoring that best suits children.
However, a systematic review in adults in Queensland hospitals revealed that monitoring of ETT cuff
pressure is routine practise in ICUs (Talekar, Udy, Boots, Lipman, & Cook, 2014), and there is strong
evidence that monitoring is crucial to avoid tracheal injuries (Jaber et al., 2006; Talekar et al., 2014).
In adult-focused literature, eight-hourly or once per-shift is the frequency adopted in practise (Labeau
et al., 2009; Rello, Lode, Cornaglia, Masterton, & Contributors, 2010). However, more recent
evidence by a prospective observational study suggests that continuous monitoring is favourable due
to the large variation of cuff pressure between patients and within patients (Memela & Gopalan,
2014). It is unclear in the paediatric context what the overall agreement is in regard to routine cuff
pressure monitoring (Bhardwaj, 2013; Tobias et al., 2012).
Ventilator circuit checks
Regularly changing the ventilator circuit prevents bacterial colonisation (Resar et al., 2014). Removal
of condensate from the ventilator is highly recommended by CDC guidelines (Klompas, Branson, et
al., 2014) as is the use of heat and moisture exchangers (HME) (Auxiliadora-Martins et al., 2012).
There are two randomised controlled trials conducted in PICUs which examine the frequency of
changing the circuit, its effects on VAP rates and cost implications. The incidence of VAP was 11.5
per 1000 ventilator days in a group of ventilated children assigned to seven-day circuit changes,
48
compared to 13.9 per 1000 ventilator days in a group assigned circuit changes every three days (p=
0.60) (Samransamruajkit et al., 2010). This randomised controlled trial also showed that alteration of
practise from three days to seven days may save up to US$22,000 per annum and decrease overall
workload.
One study by Chu et al. (2015) reported similar findings in the reduction of costs in their Neonatal
ICU. According to this study, there is no significant difference in VAP rates between changing the
ventilator circuit either every second day or weekly (8.2 versus 9.5 per 1000 ventilator days, p-
value=0.44), but less frequent changes contributed to a yearly cost saving of US$29,350. These trials
therefore support the need for ventilator circuit checks and changing at least weekly as outlined in
most paediatric VAP prevention bundles (Bigham et al., 2009; De Cristofano et al., 2016; Resar et
al., 2014; Rosenthal, Alvarez-Moreno, et al., 2012).
Sedation interruptions
Sedation in mechanically ventilated patients is routine practise to minimise patient discomfort and to
improve synchronisation between patient and ventilator (Ostermann, Keenan, Seiferling, & Sibbald,
2000). Nevertheless, sedation also increases the duration of mechanical ventilation, and hence is
associated with VAP development (Awasthi et al., 2013). In paediatrics, evidence of the effect of
daily sedation interruptions on the prevalence of VAP is limited and somewhat conflicting. A
randomised controlled trial by Gupta, Gupta, Jayashree, and Singhi (2012) compared an intervention
of daily sedation interruption midazolam bolus of 0.1mg/kg, followed by infusion, n= 46) versus a
control group of continuous infusion of midazolam; n=56 (control group). Patients in the intervention
group had daily interruption at 8.00 am, until they became responsive to verbal commands or were
agitated/uncomfortable to the extent that prompted restarting the midazolam infusion. When
recommenced, the dose of infusion was reduced by 50%. The study showed that patients with
interrupted sedation had a significantly shorter average duration of mechanical ventilation than the
patients under the continuous sedation protocol; 7.0 SD ± 4.8 days versus 10.3 ±SD 8.4 days
respectively (p=0.021). The patients in the interrupted sedation group also stayed for a shorter time
in PICU than the patients in the continuous sedation group; median 10.7 days versus 14.0 days
respectively (p=0.048) (Gupta et al., 2012).
A prospective randomised controlled trial (n=30) found that it is feasible to implement daily sedation
interruption in mechanically ventilated children (Hoog et al., 2014). In this study the intervention
group (n=15) received midazolam and morphine infusions which were stopped daily at 1 p.m. and
restarted when the COMFORT behaviour scale was ≥17. The control group (n=15) received standard
49
care of midazolam and morphine infusion and, if signs of agitation were present, bolus doses of
sedative were administered (Hoog et al., 2014). The duration of mechanical ventilation was
significantly shorter in the intervention group compared with the control group: median four days
(IQR 3.0–8.0) and nine days (IQR 4.0–10.0) respectively (p=0.03). Similarly, PICU stays were
shorter in the intervention group compared to the control group: median six days (IQR 4.0–9.0) and
10 days (IQR 7.0–15.0) respectively (p=0.010) (Hoog et al., 2014).
However, a recent multicentre randomised controlled paediatric trial comparing daily interruption
plus protocolised sedation (n=66) against protocolised sedation only (n= 63) by Vet et al. (2016)
found results that contradicted the above studies. Daily sedation interruption made no difference in
terms of the number of ventilator-free days in the daily interruption plus protocolised sedation group
(24.0 days (IQR 21.6–25.8) versus protocolised sedation only 24.0 days (IQR 20.6–26.0). There was
also no difference in median PICU length of stay with daily interruption plus protocolised sedation;
6.9 days (IQR 5.2–11.0) versus protocolised sedation only 7.4 days (IQR 5.3–12.8), p=0.47. The
length of hospital stays for daily interruption plus protocolised sedation was 13.3 days (IQR 8.6–
26.7) versus protocolised sedation only 15.7 days (IQR 9.3–33.2), p=0.19. Surprisingly, despite no
significant difference between the groups at baseline, mortality at 30 days was higher in the daily
interruption plus protocolised sedation group than in the protocolised sedation only group (6/66
versus 0/63, p=0.030) (Vet et al., 2016).
In summary, single-centred feasibility studies by Gupta et al. (2012) and Hoog et al. (2014) have
shown that the daily interruption of sedation is beneficial for critically ill children in PICU. However,
in the multicentre study, the opposite findings were obtained (Vet et al., 2016). Therefore, controversy
remains regarding the adoption of daily sedation interruption as a paediatric VAP preventative
strategy.
Early enteral feeding commencement
The role of enteral nutrition on the development of VAP is important through the prevention of
bacteria colonisation by decreasing gastric pH and ulcer occurrences (Kallet & Quinn, 2005). A meta-
analysis of data from intubated adults (n=6) demonstrated a statistically significant reduction in
pneumonia attributable to the provision of early feeding (OR=0.31, CI 0.12–0.78 p=0.010) (Doig,
Heighes, Simpson, Sweetman, & Davies, 2009).
Multiple studies recommend early commencement of enteral feeding within 24–48 hours of ICU
admission assuming that no contraindications to feeding exist (Barr, Hecht, Flavin, Khorana, &
50
Gould, 2004; Heyland, Dhaliwal, Day, Jain, & Drover, 2004). These recommendations are based on
the understanding of the positive role of early nutrition in promoting tissue repair and improving the
metabolism of the body, and consequent reduction in complications such as infection.
There is scarce evidence of the relationship between early feeding commencement and VAP
development in paediatrics. A recent randomised controlled trial involving 120 children in PICU
investigated initiation of enteral feeding — early (started within 6–24 hours) (n=60) compared with
late (started after 24 hours) (n=60) — examining the length of mechanical ventilation, PICU stay and
incidence of VAP. The study found no difference in VAP rates between patients assigned to early or
late initiation of feedings, p=0.44 (Prakash, Parameswaran, & Biswal, 2016). In an observational
study evaluating enteral nutrition initiation within the first 48 hours in 592 ventilated PICU patients
(60.0%), no association with VAP was identified (Albert et al., 2016). Overall, current evidence
shows that early feeding initiation in mechanically ventilated paediatric patients has minimal impact
in relation to the development of VAP.
Gastrointestinal (GI) prophylaxis
The effect of GI prophylaxis in prevention of VAP in children is unclear because of the low quality
of evidence and it is therefore not recommended (Klompas, Branson, et al., 2014). Despite this
recommendation, it is common practise in most PICUs to administer proton-pump inhibitors (PPIs)
or histamine-2 receptor antagonists (H2RAs) in the prevention of GI bleeding (Duffett et al., 2017;
Spirt, 2003). In a prospective observational study of 398 children from five PICUs in Brazil, 78% of
children received prophylaxis (Araujo, Vieira, & Carvalho, 2010).
A prospective cohort study by Albert et al. (2016), involving 59 PICUs from 15 countries, revealed
that the use of acid-suppressive medications (PPIs and H2RAs) in 763 patients (61.0%) increased the
likelihood of developing VAP (OR 2.0, 95% CI: 1.2-3.6, p=0.011). According to the authors, the
possible impact of acid-suppressive medications on the microbiome of the gastrointestinal tract may
contribute to this finding (Albert et al., 2016), and they suggest further examination of the rationale
for prescription of acid-suppressive drugs in PICUs.
An earlier study indicated that the use of sucralfate did not decrease the incidence of VAP in 53
children who were prescribed the sucralfate compared to 48 children who were not prescribed any
form of GI prophylaxis (Lopriore, Markhorst, & Gemke, 2002). Similarly, a prospective study by
Yildizdas, Yapicioglu, and Yilmaz (2002), revealed no difference in the incidence rate of VAP among
51
children prescribed either ranitidine, sucralfate, or omeprazole, compared with no GI prophylaxis
(p=0.96, CI= 0.95-0.96). In summary, the evidence demonstrates that the use of GI prophylaxis in
paediatric care has little impact or may even increase the risk of VAP incidence.
Summary
The implementation of the VAP prevention bundle in children has so far shown positive outcomes,
such as a reduction in VAP rates. Researchers still debate whether any individual preventative
strategy in the bundle has contributed to this reduction (Cooper & Haut, 2013; Klompas et al., 2016).
Unlike the adult VAP bundle where there are four to six well established individual preventive
strategies, the paediatric VAP bundle has individual preventative strategies which are varied and
understudied. This is likely due to lack of data available for appraising the individual VAP
preventative strategies (Cooper & Haut, 2013).
Most researchers agree that it is difficult to demonstrate high level research evidence for the ventilator
bundle with paediatrics using a randomised controlled trial (RCT) methodology (Duffett et al., 2013).
This may be partly due to feasibility issues. Some studies only analyse practises at a protocol level
due to recruitment challenges, or premature withdrawal of patients from studies subsequent to adverse
outcomes or patient deterioration (Duffett et al., 2017; Duffett et al., 2013; Vet et al., 2016).
Irrespective of these limitations, the individual VAP preventative strategies included in a bundle
should be tailored to individual patients (Klompas et al., 2016; Wip & Napolitano, 2009). This is
crucial to accommodate changes in epidemiology, treatment, diagnosis and prevention of VAP (Nair
& Niederman, 2015).
Organisational and clinician compliance with VAP preventative strategies
The implementation of VAP preventative strategies in clinical practise is hampered by an unclear
compliance benchmark. According to IHI, compliance with the bundle is generally calculated using
an all-or-none measurement rule with an aim for 95% compliance or more (Resar et al., 2014). This
means that adherence is counted either as ‘yes’ or ‘no’ for each individual preventative strategy, and
if any of the individual preventative strategies are absent, compliance is considered as incomplete
(Resar et al., 2014). A paediatric study by Bigham et al. (2009), described overall compliance for
each of the individual preventative strategies as 95% or greater, consistent with the IHI tool. However,
some researchers measure overall compliance categorically. In the latest publication by Tabaeian,
Yazdannik, and Abbasi (2017), there are four categories of compliance: (1) unacceptable compliance
at 0–25%; (2) average compliance 25–50%; (3) relatively acceptable compliance 50–75%; and 4)
acceptable compliance 75–100%.
52
Moreover, details of the level of compliance of individual VAP preventative strategies are often not
reported in publications. This may have implications on the calculation of overall compliance as a
firm compliance benchmark concerning the individual preventative strategies in children’s care is not
well defined (Bouadma, Mourvillier, Deiler, Le Corre, et al., 2010; Klompas et al., 2016; Resar et al.,
2014).
Nevertheless, there are a small number of paediatric studies, such as Bigham et al. (2009) that report
compliance of individual preventive strategies. Interestingly, they reported baseline compliance with
hand hygiene, oral hygiene, hand hygiene before and after contact with ventilator circuit, oral suction
device stored in unsealed plastic bag, in-line suction catheter changed when soiled, condensate
drained every two to four hours and when turning, ventilator circuit inspected and changed only when
soiled all at 60%. HOB elevation compliance was 57%. Following an educational program and
implementation of a paediatric VAP bundle, they reported the compliance increased to 100% for
every element. To establish if these changes were sustained they reaudited 22 months later. There
were improvements from baseline for each element with most compliance rates being 85% (Bigham
et al., 2009).
Compliance of healthcare workers to the ventilator bundle can prove to be a challenge in practise,
with poor to modest compliance being reported (Bouadma, Mourvillier, Deiler, Le Corre, et al., 2010;
Cason et al., 2007; Nair & Niederman, 2015). A recent multicentre cohort study involving five adult
ICUs reported the overall compliance with the VAP prevention bundle (inclusive of hand hygiene,
changing ventilator circuits, sedation interruption, ETT cuff pressure control, and oral care) was less
than 30.0%, with the lowest compliance being hand hygiene at 19.0% (Rello et al., 2013). Brierley et
al. (2012) reported challenges in VAP prevention bundle compliance in their PICU; however, they
managed to increase compliance from 50.0% to 90.0% after strict compliance monitoring with
monthly reporting.
Hand hygiene compliance for healthcare workers is an ongoing challenge in healthcare settings
(Allegranzi et al., 2013; Smiddy et al., 2015; World Health Organization (WHO), 2009a). The World
Health Organization (WHO) (2009b) defined non-compliance of hand hygiene as when the number
of opportunities exceeded the number of actions performed. The WHO does not provide a range of
hand hygiene compliance recommendations for patients’ families. Sickbert-Bennett et al. (2016)
classified hand hygiene compliance among healthcare workers at baseline as high level (>80.0%) or
higher level (>95.0%). The Australian Hand Hygiene benchmark for hand hygiene compliance among
healthcare workers from 2017 onwards is more than 80.0% (Hand Hygiene Australia, 2017b).
53
There are few recently published studies which evaluate hand hygiene compliance of healthcare
workers in PICUs in relation to VAP prevalence. In 2002, an interventional randomised cohort study
was conducted in a PICU/Neonatal ICU to evaluate the impact of introducing an alcohol-based hand
gel on hand hygiene compliance (Harbarth et al., 2002). This study revealed that the overall adherence
to hand hygiene was 32.0% at baseline and following the introduction of hand gel the compliance
increased to 37.0%. In two other studies, at baseline the reported hand hygiene compliance of nurses
and medical practitioners in Neonatal ICUs was only at 46% and 65% before a hand hygiene
education program was implemented and increased to 69% and 88% after the program respectively
(Chhapola & Brar, 2015; Helder, Brug, Looman, van Goudoever, & Kornelisse, 2010).
It has been noted that challenges exist in defining individual VAP preventative strategy compliance
in paediatrics, thus contributing to variation in reported compliance rates. Overall compliance
amongst healthcare workers is low to modest, with limited compliance rates reported for the
individual VAP preventative strategies. This indicates that appropriate attention needs to be given to
address these challenges and improve global compliance for all healthcare workers.
Improving compliance with VAP preventative strategies via VAP education
Since VAP and VAE are associated with poor outcomes, it is important to ensure incidence is
minimised. In addition to and consistent with VAP surveillance (Khosim et al., 2015; Marcia Regina
Eches et al., 2015), a two-pronged way of minimising VAP incidence is by VAP education
complemented by strict monitoring of the compliance (Rosenthal, Alvarez-Moreno, et al., 2012;
Smiddy et al., 2015). Current strategies include using multimodal approaches, which include
education and compliance auditing (Flodgren et al., 2013; Klompas, Branson, et al., 2014).
VAP Education
Education remains a key strategy in the prevention of infection in hospital settings (Rello et al., 2010).
A systematic review by Flodgren et al. (2013) assessed eight adult ICU studies and found that
education plays a significant role in improving patient safety. The findings of this review emphasised
the importance of active implementation of procedures via repeated lectures and VAP surveillance.
A systematic review by Borgert, Goossens, and Dongelmans (2015) ranked education (86%),
reminders (71%), and combined audits with feedback (63%) as the top three strategies for effective
implementation of VAP preventative strategies in ICUs. However, these two systematic reviews only
considered adult ICU settings.
54
In contrast, very few studies have been conducted in the paediatric setting specifically assessing the
impact of education on VAP rates (Cooper & Haut, 2013). Table 3.4 describes some important studies
in paediatric settings which acknowledge the need for updated VAP education among healthcare
workers. All of the studies mention educational approaches and interventions with Bigham et al.
(2009) and Brierley et al. (2012) presenting paediatric VAP preventative strategies in the USA and
UK respectively. These studies have been repeatedly used by other researchers as the benchmark for
VAP preventative strategies in children. Another three studies, Gupta et al. (2014), Obeid et al.
(2014), and De Cristofano et al. (2016), are more recent and strengthen the evidence on VAP
prevention and education delivery. These studies are consistent with the current medium of
information delivery (e-learning) and may encourage staff engagement with VAP education (Labeau
et al., 2016; Reime, Harris, Aksnes, & Mikkelsen, 2008). All studies demonstrate a positive
association between VAP education and the reduction of VAP rates in PICU (Table 3.4).
A challenge in ensuring compliance with individual VAP preventative strategies is maintaining active
practise over time (Kollef, 2011). Education is essential to maintain a high level of overall compliance
with a VAP prevention bundle (Bigham et al., 2009; Brierley et al., 2012; De Cristofano et al., 2016;
Rello et al., 2013). A quasi-experimental study in a neonatal setting revealed the positive impact of
an education intervention on hand hygiene compliance among healthcare workers indicating that
education is paramount in addressing compliance issues (Chhapola & Brar, 2015). The education
included training sessions, reminders and auditing and feedback. After implementation of these
programs, compliance of hand hygiene increased from 46% to 69% (Chhapola & Brar, 2015). An
earlier study by Helder et al. (2010) also demonstrated an increase in hand hygiene compliance
amongst nurses and other healthcare workers from 65% to 88% after the implementation of education
programs and feedback.
55
Table 3.4: Paediatric studies involving VAP education and VAP preventative strategy implementation
Authors/country Design/participants Intervention/ approach Main outcome
measures
Results
Bigham et al.
(2009)/USA
Pre- post
implementation
design; Process-
Improvement
Initiative/ PICU
healthcare workers
(i) VAP bundle education and implementation of
paediatric-VAP bundle (ii) New mouthcare product
(iii) Compliance auditing, (iv) Modify policy and (v)
Advertising latest VAP rates via posters
(i) VAP rate
(ii) PICU
length of stay,
duration of
ventilation,
(iii) Mortality
(i)The VAP rates reduced from 5.6 at
baseline to 0.3 infections per 1000
ventilator days
ii) A significant association found
between VAP and length of stay:
VAP 19.5 days SD ± 15.0 vs non-
VAP 7.5 days ± 9.2, p < 0.001,
duration on ventilator: VAP 16.3 days
SD ± 14.7 vs non-VAP 5.3 days SD ±
8.4, p < .001
(iii) Mortality: VAP 19.1% vs non-
VAP 7.2%, p= 0.010)
Brierley et al.
(2012)/UK
Pre- post
implementation
design; Quality
improvement
study/healthcare
workers
A nurse-led VAP surveillance programme involved (i)
Multiple training opportunity, (ii) VAP education as a
routine part of staff induction, (iii) Compliance
monitoring, added to existing implementation of VAP
bundle and surveillance in the unit
(i)VAP rate
(ii)Compliance
to VAP bundle
(i) The VAP rates reduced from 5.6
to 0.3 per 1000 ventilator days
(ii) Compliance to VAP preventative
strategies progressively increased
Gupta et al.
(2014)/India
Prospective
cohort/nursing staff
and resident doctors
Education program delivered through lectures on (i)
Introduction to HAIs, hand hygiene, standard
precautions, (ii) Standard operating procedure for
sample collection, (iii) Pathophysiology of VAP and
VAP prevention strategies, sterilisation and
disinfection of respiratory equipment, aseptic
procedures for suctioning of ventilated patients.
Education materials displayed in the unit (poster of ‘5
moments of hand hygiene’), added to the existing VAP
bundle implementation in the unit
(i) Incidence
of VAP
(ii) Mortality
(i) The incidence density of VAP
was reduced by 28%
(ii) Pre-intervention=20.2 and post
intervention=14.6 per 1,000
ventilator days, p= 0.21)
(iii) A statistically significant
reduction in mortality among patients
who received mechanical ventilation
for ≥ 48 hours in the post
intervention period (49.3%
vs 31.4%: p= 0.029
56
Obeid et al.
(2014)/Lebanon
Pre- post-
interventional trial
/PICU healthcare
workers
(i) Education on VAP via presentations and discussion
on VAP elements, (ii) Development of a VAP bundle
checklist, (iii) Compliance auditing daily with
feedback
(i) VAP rate
(ii) duration of
mechanical
ventilation
(i) 52 to 6 per 100 ventilator days
(ii) Patient with VAP spent longer
duration on ventilator mean 11.42
days versus 5.18 days (p< 0.0001)
De Cristofano et
al. (2016)/
Argentina
Quasi-experimental
uninterrupted time
series/PICU
healthcare workers
(i) Launched a VAP prevention program. Staff notified
via email and daily presentations by the PICU
leadership team. (ii) Developed VAP education
material including text documents, slide presentations,
posters, and videos. (iii) Displayed educational
material in the PICU virtual campus highlighting the
VAP rate.
(i) VAP
incidence
VAP rate reduction from 6.3, 5.7,
3.2, 1.8 and to 0.0
57
The evidence indicates that VAP education is essential to not only help maximise the compliance of
VAP preventative strategies but also to reduce VAP incidence rates (Haut, 2015; Klompas, Branson,
et al., 2014; Singh, Kumar, Sundaram, Kanjilal, & Nair, 2012). VAP education delivered through a
multimodal approach helps to engage healthcare workers in the PICU environment.
Despite the lack of evidence concerning which educational interventions are most effective to foster
compliance with VAP preventative strategies, Flodgren et al. (2013) suggest in their systematic
review that multiple educational interventions that are periodically implemented in the unit will best
increase adherence amongst healthcare workers. This review also identified that VAP education
exclusively targeting healthcare workers often ignores the potential for educating parents and family
members in hand hygiene promotion and practise (World Health Organization (WHO), 2009a).
The role of parents in VAP prevention and parental education on hand hygiene
The presence or involvement of parents as part of the overall healthcare team is supported (Leape et
al., 2009; Piper, 2011); it is associated with an increase in family and parental satisfaction (Voos &
Park, 2014). Parents have a strong desire to be acknowledged and actively involved in the care of
their child (Meert, Clark & Eggly., 2013). Flexible visiting hours in PICU support parental
availability at the bedside and increase the potential for parent/patient contact and parent/healthcare
worker encounters (Bellissimo-Rodrigues et al., 2016). In PICU perspective, parents’ involvement
is very limited given the complexities of care that often involve complicated medical advancement
and the venerability of children receiving care (Stickney, Ziniel, Brett, & Truog., 2014). In VAP
preventative strategies implementation, they presumably are not handling ventilator tubing,
participating in endotracheal suctioning and oral hygiene. However, directly or indirectly, parents are
well positioned to interact with their child (physical contact i.e. touching and caressing) (Bellissimo-
Rodrigues et al., 2016) and become an observer to healthcare workers’ practises for example
ventilator care bundle implementation.
Hand hygiene stands out as the primary method in breaking the spread of infection (World Health
Organisation (WHO), 2009b; Kozlowski, 2012). In the context of VAP, the cross transmission of
VAP causative organisms to the patients often associated with the issue of contamination, bacterial
colonisation through the direct contact from healthcare providers, partly due to poor hand hygiene
practice (Safdar et al., 2005; Craven, 2006). This could be due to constant challenges in hand hygiene
compliance among healthcare workers (Belela-Anacleto et al., 2018; Erasmus et al., 2010). It is also
a challenging situation where the expectation for parents and visitors to perform good hand hygiene
practice in healthcare settings as this is the primary measures to combat with HCAIs (Anthony et al.
58
(2013); World Health Organisation (WHO), 2009b.). Yet little is known about educating parents and
families in PICUs, particularly on hand hygiene.
Education on the importance of hand hygiene in PICU is also vital not only to healthcare workers,
but also for parents and visitors (Belela-Anacleto et al., 2018; World Health Organisation, 2009a).
Wu et al. (2013) showed that patients/family members who had experience with HCAIs were more
likely to remind healthcare workers to perform hand hygiene (78.4%). Chen and Chiang (2007)
evaluated the effectiveness of a hand hygiene teaching program for families of children in a PICU via
a quasi-experimental time series. The experiment group received an introduction to hand hygiene
from nursing staff. The parents were taught by nursing staff and through watching a video about how
and when to perform hand hygiene. Families in the control group also received the education by
nursing staff and had a reference poster. The results revealed that families in the experimental group
had a higher compliance level compared to families in the control group.
Another paper in a Neonatal ICU, Anthony et al. (2013), reports on practise guidelines and parental
inclusion in hand hygiene practise to reduce Gram-negative bacterial infections. Attention was given
to educating parents on good hand hygiene practise. A more recent quality improvement study, by
Chandonnet et al. (2017), examined hand hygiene compliance among parents and family members in
a Neonatal ICU. The project was initiated due to low compliance with hand hygiene amongst these
visitors and aimed to achieve 100% compliance. The educational materials consisted of (1)
educational sheets about hand hygiene in multiple languages; (2) hand hygiene posters and stickers;
(3) real-time feedback, visual reminders of compliance reports and availability of supplies for hand
hygiene (hand rub, soap etc.). The compliance with proper hand hygiene that was observed among
parents and family members after the education program increased to 89% from 71% at baseline.
In this study, incorporating hand hygiene through parents’ participation to prevent VAP was to ensure
they have cleaned their hands (five moments) before in contact with their ventilated child, also parents
act as hand hygiene promoter when they are observing healthcare workers performing ventilator care
bundle to prevent the VAP.
Perceptions of ‘Speaking up for hand hygiene’ among healthcare workers and parents
The concept of speaking up for patient safety is an emerging area of interest and synonymous with
prevention of medication errors and promotion of hand hygiene practise in healthcare settings
(Daniels et al., 2012; World Health Organization (WHO), 2009a). The expectation of this initiative
is to provide immediate feedback to prevent human errors before harm occurs (Okuyama, Wagner,
59
& Bijnen, 2014). Despite the potential benefits in error prevention, issues arise as to what extent
healthcare workers, patients and families/parents engage with and respond to this initiative (Bsharat
& Drach-Zahavy, 2017).
Perceptions of speaking up may vary, based on: (1) the context of ‘Speak Up’™ (i.e., medication
error, hand hygiene), (2) the roles (i.e., healthcare workers, patients, families/parents), and (3) the
settings (i.e., adult, paediatric/neonatal). These factors could become barriers for collaboration
(Bellissimo-Rodrigues et al., 2016; Bsharat & Drach-Zahavy, 2017). A study that explored the
perception of patients and their involvement in speaking up for safety found that patients had a
positive experience. The healthcare workers respected the patients’ involvement in the prevention of
medication errors; however, they objected to reminders about hand hygiene (Schwappach, Frank, &
Davis, 2013). According to a systematic review by (Okuyama et al., 2014) hesitation to speak up
among healthcare workers can be a main factor contributing to communication errors.
When the ‘Speak Up’™ initiative was first introduced in 2002, the target of the initiative was the
patient themselves, emphasising their role in promoting their own safety (The Joint Commission,
2018). It is an initiative that uses easily understood materials such as leaflets and videos on evidence-
based clinical practises (e.g. hand hygiene; ‘Speak Up: prevent errors’). Later, a similar approach was
extended to family/caregivers of the patients. Evidence of the impact of the extended speak up
initiative that the research lacks rigour (Bsharat & Drach-Zahavy, 2017).
A survey in an adult hospital revealed that 75% of patients/families (n=334) were willing to remind
healthcare workers to clean their hands. In the same survey only 31% of nurses and 26% of medical
practitioners would welcome reminders from patients/families (Kim et al., 2015). The families of
patients hospitalised in the paediatric department were more reluctant to remind healthcare workers
about hand hygiene (OR 1.95; 95% CI: 0.99-3.83, p=0.053) but agreed that they could help to remind
healthcare workers about hand hygiene practise (96.5%; 333/345) (Pan et al., 2013). The intention of
patients/families to remind healthcare workers to perform hand hygiene only rated at 67.2% (232/345)
(Pan et al., 2013). An interesting finding by Wu et al. (2013) showed that patients/family members
who had experience with HCAIs were more likely to remind healthcare workers to perform hand
hygiene (78.4%). If the patient/family felt comfortable, they were increasingly willing to ask the
nurses and doctors to clean their hands (nurse: from 50.8% to 76.3%, p<0.001; doctor: from 48.9%
to 74.6% p<0.001).
60
In a large study in a tertiary hospital in Taiwan, 63% (553/880) of healthcare workers were willing to
remind patients/families about hand hygiene (Pan et al., 2013). This survey included 345
patients/families (115 patients, 220 family members and 10 attendants) and 880 healthcare workers
(241 medical practitioner, 505 nurses, 69 medical/nursing students and 65 technicians). The survey
also found that healthcare workers would feel ashamed if they were reminded by the patients/families
to practise hand hygiene. This study recommended that a mutual communication method should be
developed between staff, patients and families on hand hygiene to empower patients and families.
Illiterate families are more reluctant to remind healthcare workers to perform hand hygiene (OR, 1.79;
95% CI: 1.08-2.94, p=0.025) (Pan et al., 2013). Hierarchies within the various healthcare professions
also influence confidence levels in prompting colleagues to perform hand hygiene. For example,
junior staff reminding their senior work colleagues to perform hand hygiene is frowned upon in many
settings (Kobayashi et al., 2006; Samuel et al., 2012).
The role of patients and their families in speaking up is subtly different when the patient is a child
and the family members are parents. A systematic review by Bellissimo-Rodrigues et al. (2016)
examined 11 papers on the role of parents in the promotion of hand hygiene in paediatric centres. The
results suggest that parents understand the importance of hand hygiene to prevent infection, but they
lacked knowledge on hand hygiene procedures.
According to Coyne, Murphy, Costello, O’Neill, and Donnellan (2013) and Smith, Swallow, and
Coyne (2015), parents/family members are highly concerned about the attitudes of healthcare workers
towards their involvement with safety issues. Most parents and healthcare workers are aware that
hand hygiene is important for the prevention of infection in hospital (Bellissimo-Rodrigues et al.,
2016; Wu et al., 2013), but the willingness to remind each other of hand hygiene in practise varies
considerably. Similar research by Ciofi degli Atti et al. (2011) found that 56% of parents did not
believe that they should remind healthcare workers about hand hygiene. For their part, many
healthcare workers perceived that a reminder from parents to perform hand hygiene was unnecessary
(48.9%) (Ciofi degli Atti et al., 2011).
The literature demonstrates that the perception of ‘Speaking up for hand hygiene’ is equally valued
by parents and healthcare workers. Both parties agree that the initiative is important and helps prevent
infection, but participants were not always confident or comfortable in reminding each other, with
more healthcare workers preferring minimal participation from parents/family members. There is
minimal study on ‘Speaking up for hand hygiene’ specifically in the VAP prevention among parents
in PICU. Hand hygiene compliance monitoring amongst parents and visitors is not mandatory in most
61
institutions, unlike hand hygiene compliance among healthcare workers (Hand Hygiene Australia,
2017a; World Health Organization (WHO), 2009a). This lack of surveillance may underestimate the
demand for education on hand hygiene for parents and visitors (Giannini et al., 2016).
VAE and its preventative strategies
VAP preventative strategies and the challenges in implementation have been comprehensively
discussed in this chapter. For VAE, however, little is known about the specific preventative strategies
in paediatric and adult settings. Available evidence in adult populations suggests that VAE
preventative strategies are based on associations with pneumonia, fluid overload, atelectasis and
ARDS (Hayashi et al., 2013; Klompas et al., 2015). Thus, avoiding intubation, sedation interruption,
minimising the use of sedation (Lewis et al., 2014), paired daily spontaneous awakening and
breathing trials (Muscedere et al., 2013), and low tidal volume ventilation (Neto et al., 2015) are
among the VAE preventative strategies recommended. In paediatric literature, there are no studies
which examine VAE preventative strategies.
Summary
This chapter describes VAP preventative strategies that when combined are referred to as a VAP
bundle. The variation of individual VAP preventative strategies in a bundle in paediatrics is noted.
This review indicates that some individual VAP preventative strategies in adult VAP bundles are
applicable to paediatrics, with specific modification of strategies such as HOB elevation and cuff
pressure value. There is less evidence to support the use of GI prophylaxis and sedation interruptions.
The challenges, especially with compliance with VAP preventative strategy implementation, continue
in clinical settings. The review also indicates that hand hygiene is a crucial element/individual
preventative strategy, and this highlights the necessity of updating VAP education among PICU staff.
This review also acknowledges the parental role in a clinical setting and the value of educating parents
on ‘Speaking up for hand hygiene’ which may add new insight to VAP prevention in PICU.
The next chapter describes the methodology of this study.
62
Research Methodology
Introduction
This chapter outlines the approaches used in the research described in this thesis. There were three
phases to this work: Phase 1: Retrospective study; Phase 2: VAP education with VAP preventative
strategies compliance auditing and cross-sectional surveys; and Phase 3: Prospective study (Figure
4.1). This study utilised quantitative approaches with a small qualitative component in the Phase 2
cross-sectional surveys.
Figure 4:1: Study timeframe with respective research activities undertaken
Study setting
The study was carried out in the paediatric intensive care unit (PICU) of Queensland Children’s
Hospital (QCH) formerly known as Lady Cilento Children’s Hospital (LCCH). The change of name
to QCH was effective from 21st September 2018. This metropolitan hospital is a large and modern
dedicated paediatric facility in Australia with a capacity of 35 ICU beds (Children's Health
Queensland, 2016).
Study design
An epidemiological approach was adopted for early phases of the research described in this thesis.
Epidemiology is “the study of the distribution and the determinants of health-related states or events
in specified populations and the application of this study to control health problems” (Porta et al.,
2014, p. 13). This study involved two major epidemiological study designs: descriptive or
Phase 1:
Retrospective data collection
Mid-Nov 2016 to early Mar 2017
Phase 2:
VAP Education, VAP preventative
strategies' compliance auditing
and, Surveys
Mar to May 2017
i) VAP Education (PICU staff and Parents)
ii) VAP preventative strategies' compliance auditing with feedback
iii) Parental and Nurses surveys
Phase 3:
Prospective data collection
June to Dec 2017
63
observational and interventional (Woodward, 2014). The descriptive study design allows
identification of the status of the problem over time, which permits an analytical process to focus on
the possible risk factors associated with the disease occurrence. The interventional study design
involves testing the preventative measures or evaluation of programs/strategies that have been
implemented (Webb, Bain, & Page, 2017).
Phase 1: Retrospective study was a descriptive study which aims to summarise the prevalence of
VAP and VAE at baseline. Phase 2: VAP education used analytical study designs with quantitative
and qualitative methods and includes compliance auditing with interim feedback and a cross-sectional
survey of both parents and PICU staff. To determine the influence of VAP education and compliance
auditing with feedback, Phase 3: Prospective study used an interventional study design.
The prevalence of VAP and VAE and compliance with VAP preventative strategies was assessed in
Phase 1. An updated VAP education program and VAP compliance auditing inclusive of feedback
were introduced in Phase 2 and the impact of these were evaluated in Phase 3. These approaches
enabled the research questions to be adequately examined in this thesis.
Phase 1: Retrospective study
The purpose of the retrospective study was to assess the current status of VAP and VAE in a paediatric
population using the PNU1/VAP surveillance tool and the VAE surveillance tool. Retrospective
studies are widely used in epidemiological studies (Jansen et al., 2005) for evaluation of quality
improvement studies (Allison et al., 2000) and assessment of inpatient care (Ashton, Del Junco,
Souchek, Wray, & Mansyur, 1997). Retrospective studies allow analysis of a large dataset with less
cost (Clark, 2008) and are able to provide robust, naturalistic data to inform the evaluation of
treatment patterns and clinical outcomes and safety (Stein, Bassel, & Payne, 2014). The design is
indeed useful in analytical techniques for patient safety studies (Weinger, Slagle, Jain, & Ordonez,
2003).
In the retrospective study there were four research questions to be addressed:
1. What was the incidence and prevalence of VAP and VAE in QCH PICU in 2015?
2. What was the sensitivity and specificity of the VAE surveillance tool compared to the PNU1/
VAP tool?
3. What were the risk factors and compliance with VAP preventative strategies in QCH PICU
in 2015?
64
4. What were the risk factors of VAP and VAE and preventative strategies associated with a
diagnosis of VAP and VAE in QCH PICU in 2015?
The specific approaches employed in the methodology for this phase respond directly to the research
questions above.
Population and sample
All patients admitted to QCH PICU requiring invasive mechanical ventilation for greater than or
equal to 48 hours from 1 January 2015 to 31 December 2015 were included in this phase. The
surveillance involved all admissions to QCH PICU.
The inclusion criteria were;
1. age 0 to 18 years old;
2. invasive mechanically ventilated for ≥48 hours;
3. intubated with an endotracheal tube (ETT).
The exclusion criteria were:
1. non-invasive ventilation;
2. ventilated for <48 hours;
3. admission with existing tracheostomy intubation.
Data collection
The data were obtained from the Centre for Children’s Health Research at South Brisbane, through
the Children’s Health Queensland servers between mid-November 2016 and early March 2017. Data
was entered into a Microsoft Excel spreadsheet. The electronic repositories involved were Metavision
(iMDsoft®), Enterprise Picture Archive and Communication System (PACS), AUSLAB (Pathology
Queensland) and Hospital Based Corporate Information System (HBCIS). Monthly hand hygiene
compliance data of PICU nurses was obtained from Patient Safety and Quality Service, Children’s
Health Queensland (Children’s Health Queensland Hospital and Health Service, 2016). The data
derived from audit process undertaken monthly involving trained observers watch for opportunities
(5 moments) from a sample of nurses in PICU. A pilot study was undertaken to maximise the
efficiency of data retrieval and confirmation validity of data using sets of dummy data combined with
frequent communication with other researchers who had undertaken similar work. Appendix A was
the data collection sheet used in the Phase 1 study.
65
VAP and VAE surveillance tools
The PNU1/VAP surveillance tool
This tool was previously described in Chapter 2 (see Table 2.2).
The VAE surveillance tool
The VAE surveillance tool was previously described in Chapter 2 (see Table 2.3).
Demographic characteristics and admission-related variables
The choice of demographic characteristics was based on a review of the literature and the guidance
provided by clinical colleagues working in the area. Each of the variables collected were defined in
Appendix A, including their units of measurement for continuous and categorical variables.
Potential risk factors
A review of the literature was undertaken to identify potential risk factors for both VAP and VAE in
the paediatric population. These include the most frequently reported risk factors found to be
associated with VAP development in paediatrics such as reintubation, the presence of paralytic
agents, sedation level, nasogastric tube, absence of gastrointestinal prophylaxis and routes of
intubation (Casado et al., 2011; Gautam et al., 2012; Kusahara et al., 2014; Liu et al., 2013). Each of
the risk factors was defined in Appendix A.
VAP preventative strategies
The PICU at QCH had pre-existing documented procedures in place for VAP preventative strategies,
known as the VAP bundle. The bundle consisted of seven preventative strategies: hand hygiene, oral
hygiene, endotracheal suctioning (open suction), head of bed elevation, cuff pressure checks,
ventilator circuit checks and initiation of enteral feeding within 24 hours of PICU admission
(Queensland Children's Hospital Paediatric Intensive Care Unit, 2016). The VAP bundle was
developed at the Mater Children’s Hospital and transitioned to its existing form at QCH in early 2015.
The frequency of each VAP preventative strategy performance within a 24-hour period was recorded
and the variables were defined in Appendix A. The hand hygiene data derived from audit process
undertaken monthly involving trained observers watch for opportunities (5 moments) from a sample
of nurses in PICU. This validated hand hygiene data was derived from Patient Safety and Quality
Service, Children’s Health Queensland (Children’s Health Queensland Hospital and Health Service,
2016).
66
Data analysis
Data were analysed using the Statistical Package for the Social Sciences software IBM version 24,
(IBM, 2016) and a general-purpose statistical software, STATA® version 14 for multivariate analysis
(STATA 14, StatCorp, College Station, TX). The data in Excel format were checked and exported
into Statistical Package for the Social Sciences software. Data checking, and cleaning processes were
performed. Data were checked and cleaned by running frequencies, calculating maximum, minimum,
mean, standard deviation (SD), median, interquartile range (IQR) and using contingency tables. This
ensured that the data were free from errors and anomalies, thus permitting analysis of the correct data
to answer the research questions (IBM Knowledge Center, 2011). Univariate analysis of demographic
data, risk factors and preventative strategies required use of the Student t-test or the Mann-Whitney
U test for non-normally distributed continuous variables, Chi-squared test or Fisher’s exact test where
more than 20% of the expected counts are less than 5 for categorical variables. The significance level
used was 0.05.
The experimental unit was defined as episodes of mechanical ventilation; however, data are
summarised by patient and admission where appropriate, and clearly indicated as such. Results are
presented as frequency and percentages for categorical data, and means and SD for normally
distributed continuous variables, and medians and IQR for non-normally distributed continuous
variables.
4.3.1.7.1 Incidence of VAP and VAE
The outcome of interest was a binary event and the calculation of the incidences of VAP and VAE
were undertaken separately. Firstly, VAP events were identified after fulfilment of the PNU1/VAP
surveillance tool and confirmed by a clinician from the unit and research supervisor. VAE events
were identified after fulfilment of the VAE surveillance tool and classification was then confirmed
using the VAE Calculator Version 4.0 from the CDC and NHSN website (Centers for Disease Control
and Prevention (CDC), 2016b).
The first incidence rate was calculated using the CDC formula which sums all ventilation days within
the study period.
The following formula was used:
𝐼𝑛𝑐𝑖𝑑𝑒𝑛𝑐𝑒 𝑟𝑎𝑡𝑒 =𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑛𝑒𝑤 𝑉𝐴𝑃 𝑜𝑟 𝑉𝐴𝐸 𝑑𝑖𝑎𝑔𝑛𝑜𝑠𝑒𝑑 𝑑𝑢𝑟𝑖𝑛𝑔 𝑡ℎ𝑒 𝑠𝑡𝑢𝑑𝑦 𝑝𝑒𝑟𝑖𝑜𝑑
𝑆𝑢𝑚 𝑜𝑓 𝑜𝑏𝑠𝑒𝑟𝑣𝑒𝑑 𝑣𝑒𝑛𝑡𝑖𝑙𝑎𝑡𝑖𝑜𝑛 𝑑𝑎𝑦𝑠 𝑑𝑢𝑟𝑖𝑛𝑔 𝑡ℎ𝑒 𝑠𝑡𝑢𝑑𝑦 𝑝𝑒𝑟𝑖𝑜𝑑
67
The second incidence rate was calculated to adjust for time at risk for developing VAP/VAE, which
is not taken into consideration within the CDC calculation. The endpoint for observed ventilation
days were defined as either:
• until ventilation is ceased in the absence of the development of VAP or VAE;
• until diagnoses of VAP or VAE; or
• death.
The incidence of VAP or VAE was then expressed per 1000 ventilation days.
4.3.1.7.2 Diagnostic accuracy of the VAE surveillance tool
The diagnostic accuracy of the VAE surveillance tool compared to the PNU1/ VAP tool was
determined using the following formulae:
a) 𝑆𝑒𝑛𝑠𝑖𝑡𝑖𝑣𝑖𝑡𝑦 =𝑡𝑟𝑢𝑒 𝑝𝑜𝑠𝑖𝑡𝑖𝑣𝑒
𝑡𝑟𝑢𝑒 𝑝𝑜𝑠𝑖𝑡𝑖𝑣𝑒 +𝑓𝑎𝑙𝑠𝑒 𝑛𝑒𝑔𝑎𝑡𝑖𝑣𝑒
b) 𝑆𝑝𝑒𝑐𝑖𝑓𝑖𝑐𝑖𝑡𝑦 =𝑡𝑟𝑢𝑒 𝑛𝑒𝑔𝑎𝑡𝑖𝑣𝑒
𝑓𝑎𝑙𝑠𝑒 𝑝𝑜𝑠𝑖𝑡𝑖𝑣𝑒 + 𝑡𝑟𝑢𝑒 𝑛𝑒𝑔𝑎𝑡𝑖𝑣𝑒
c) 𝑃𝑜𝑠𝑖𝑡𝑖𝑣𝑒 𝑃𝑟𝑒𝑑𝑖𝑐𝑡𝑖𝑣𝑒 𝑣𝑎𝑙𝑢𝑒 =𝑡𝑟𝑢𝑒 𝑝𝑜𝑠𝑖𝑡𝑖𝑣𝑒
𝑡𝑟𝑢𝑒 𝑝𝑜𝑠𝑖𝑡𝑖𝑣𝑒+𝑓𝑎𝑙𝑠𝑒 𝑝𝑜𝑠𝑖𝑡𝑖𝑣𝑒
d) 𝑁𝑒𝑔𝑎𝑡𝑖𝑣𝑒 𝑃𝑟𝑒𝑑𝑖𝑐𝑡𝑖𝑣𝑒 𝑣𝑎𝑙𝑢𝑒 =𝑡𝑟𝑢𝑒 𝑛𝑒𝑔𝑎𝑡𝑖𝑣𝑒
𝑓𝑎𝑙𝑠𝑒 𝑛𝑒𝑔𝑎𝑡𝑖𝑣𝑒+𝑡𝑟𝑢𝑒 𝑛𝑒𝑔𝑎𝑡𝑖𝑣𝑒
Values were calculated by hand and then entered into the online calculator:
https://www.medcalc.org/calc/diagnostic_test.php to calculate the 95% confidence intervals
(Altman & Bland, 1994; MEDCALC easy-to-use statistical software, 2017).
4.3.1.7.3 Assessment of potential risk factors and preventative strategies for VAP and VAE
Assessment of potential risk factors and preventative strategies in this study employed a multivariate
analytical approach using Poisson and Negative Binominal Regression models. All of the final
models were run in STATA (STATA 14, StatCorp, College Station, TX).
Variables used in the multivariable modelling
Response variables
1. Ventilator-associated pneumonia (VAP): Yes or No during a single episode of mechanical
ventilation.
2. Ventilator-associated event (VAE): Yes (defined as VAC, IVAC or PVAP) or No during a
single episode of mechanical ventilation.
Exposure variables
Hours of at risk was defined by case classification as:
68
1. For those cases that were classified as having VAP: time of first x-ray found to be positive
(≥48 hours) minus time mechanical ventilation commenced.
2. For those cases that were not classified as having VAP: time mechanical ventilation
ceased/patient died minus time mechanical ventilation commenced.
3. For those cases that were classified as having VAE: the time of worsening oxygenation (≥48
hours/2 calendar days) minus time mechanical ventilation commenced.
4. For those cases that were not classified as having VAE: time mechanical ventilation
ceased/patient died minus time mechanical ventilation commenced.
Potential explanatory variables
• Gender: (male, female)
• Age: (months)
• Weight: (kg)
• Reintubation episodes (yes, no)
• Underlying disease (categorical: trauma/injury, cardiovascular, neurological, respiratory,
renal, gastrointestinal, infection and miscellaneous)
• Paralytic agent (yes, no)
• Gastrointestinal prophylaxis (yes, no)
• Nasogastric tube presence (yes, no)
• Routes of intubation: (categorical: nasal, oral)
• Sedation level: (categorical: deep sedation, light sedation, agitated)
• Percentage of compliance in month of hand hygiene (continuous)
• Average of frequency oral hygiene performed per total mechanical ventilation days
(continuous)
• Average of frequency endotracheal suctioning performed (open suction)/total mechanical
ventilation days (continuous)
• Average of frequency head of bed elevation performed/total mechanical ventilation days
(continuous)
• Average of frequency cuff pressure checks performed/total mechanical ventilation days
(continuous)
• Average of frequency ventilator circuits checks performed/total mechanical ventilation days
(continuous).
69
The Poisson regression was run initially as it allows assessment of the time that patients were at risk
for VAP and VAE, where the patients are followed over a variable length of time. This accounts for
the adjustment required to exclude children that had already developed VAP/VAE. These children
were no longer at risk for developing VAP/VAE so were then not included in further at-risk
calculations.
With a very low VAP and VAE event, logistic regression was not appropriate as it did not meet the
assumptions; the patients were not at risk for the same time period and the event rate was less than
10%, which would result in an underestimation of the probability of the rare events (King & Zeng,
2001). Since the study outcome is in binary form, a robust error variance procedure called sandwich
estimation was required to correct for overestimation of the error for the estimate parameter in the
Poisson regression model (Zou, 2004). This type of Poisson model is referred to as a modified Poisson
model.
The modified Poisson model has the following probability distribution:
𝑓(𝑦; 𝜇) =𝜇 𝑦𝑒−𝜇
𝑦!, 𝑦 = 0,1
Where:
y is the number of VAP or VAE (yes/no)
𝜇 is the average rate of VAP or VAE per hour of ventilation.
The present study used the mechanical ventilation episode as the experimental unit and each
mechanical ventilation had a minimum duration of 48 hours. The VAP/VAE event is recorded for
each mechanical ventilation. The dataset also counted mechanical ventilation episodes within
admissions, and each admission is nested within patients, but the effect of correlation present between
mechanical ventilation within an admission is unknown. Despite this, Elward et al. (2002) used
mechanical ventilation as the experimental unit in their study. To examine the effect of correlation
between mechanical ventilation within an admission, a mixed effects Poisson regression model was
carried out using the three-level, random-intercept Poisson model described as:
log(𝜇𝑖𝑗𝑘) = log(expected𝑖𝑗𝑘) + 𝛽0 + 𝛽1X𝑖𝑗𝑘 … . +𝑢𝑘 + 𝑣𝑗𝑘
Where mechanical ventilation within an admission is j, subjects k, and mechanical ventilation
episodes i, X is an explanatory variable. 𝛽0 is the intercept estimate, 𝛽1is the estimate for a predictor,
70
𝑢𝑘 is random patient effect and 𝑣𝑗𝑘 is the random mechanical ventilation within an admission effect
nested within the patient. The model includes the offset term for time at risk, to obtain incidence rate
ratio per hour of ventilation. A sandwich estimator was again applied. The likelihood ratio test is used
to examine whether the mixed effects model better explains the data than the modified Poisson.
When the Poisson models are over dispersed, it means that the assumption of modified Poisson
regression is not met (the mean does not equal the variance). An alternative is the Negative Binomial
regression models which were performed using STATA version 14 (STATA 14, StatCorp, College
Station, TX). An offset term for time at risk was also examined. To correct the underestimate, the
standard errors of estimates, a gamma-Poisson distribution in addition to the dispersion parameter
relaxes the strict assumption of the Poisson model by accounting for heterogeneity. These, in theory,
should perform similarly with a sandwich estimator as per the modified Poisson model since to the
best of our knowledge, no other study that we could find has used a Negative Binominal with a binary
outcome.
A Negative Binomial model has the following probability distribution:
𝑓(𝑦; 𝑟, 𝑝) = (𝑦 + 𝑟 − 1
𝑦) 𝑝𝑦(1 − 𝑝)𝑟, 𝑦 = 0,1
Where:
y is VAP/VAE (yes or no)
r is the number of cases without VAP/VAE
p is the probability of VAP/VAE.
A three-level random-intercept negative binominal model is as per the three-level random-intercept
Poisson model except it contains the additional dispersion parameter.
Steps to the univariate and multivariate analysis:
Step 1: Understand the variables under consideration. From a clinical perspective, is it
biologically plausible that they could be a predictor of VAP or VAE? What are the potential
relationships between risk factors and between risk factors and demographics? What form of
each variable would be most interpretable?
Step 2: Examine at the number of events.
Step 3: Conduct univariate analysis to explore the presence of an association between each
potential risk factor and the VAP and VAE response variables (described above), taking note of
small cell frequencies and empty cells.
71
Step 4: Conduct univariate analysis for each risk factor with other risk factors to assess for
collinearity.
Step 5: Run age- and gender-adjusted Poisson model with robust estimator for one predictor at
a time and report the adjusted incidence risk ratio.
Step 6: For those variables with a p-value of less than or equal to 0.15 in the age- and sex adjusted
Poisson models, step-wise backwards selection was used to identify the final predictors in the
model with p-value≤0.05.
Step 7: Conduct diagnostics of model to ensure model fits assumptions [over-dispersion=square
root (Pearson statistic/df, linearity of continuous variable on the logit using squared terms and
Box-Tidwell transformation), examine overall fit (Deviance goodness-of-fit and Pearson
goodness-of-fit, AIC, BIC, log-likelihood) and influence of any outliers (Pearson and Deviance
residuals)].
Step 8: Repeat analysis for the negative binomial model and the mixed effects Poisson and
Negative Binomial models.
Assessment of compliance of VAP preventative strategies
Data relating to preventative strategies were further analysed using descriptive and univariate analysis
with comparison to the compliance standards in the PICU VAP guidelines (Table 4.1).
Table 4.1: Compliance standard for VAP preventative strategies according to PICU and national
standard for hand hygiene
VAP Preventative Strategies PICU/ national standard
1. Hand hygiene > 80%
2. Oral hygiene 6 times/24 hours
Adhered to 12-hourly oral health
assessment
Twice/24 hours
Adhered to age-appropriate oral
hygiene guideline
i) Child under 6 months old without teeth
Moistening (4 hourly or 6 times)
Pink swab with sterile water (0800 & 2000; 1200 & 2400; 0400 & 1600)
ii) Above 6 months with teeth
Tooth brush (twice) - Tooth paste (0800 & 2000)
Mouth rinse (twice) - Chlorhexidine (1200 & 2400)
Moistening (twice) - Pink swab with sterile water (0400 & 1600)
3. Endotracheal suctioning (open
suction)
4 times/ 24 hours
4. Head of bed elevation 24 times/24 hours
72
5. Cuff pressure checks Twice/24 hours
Adhered to 12 hourly cuff pressure
checks
Maintained cuff pressure readings
within the limit
Min 10cmH20 to max 20cmH20
6. Ventilator circuits checks 24 times/24 hours
7. Enteral feeding commencement
within 24 hours of PICU
admission
Started within 24 hours of admission if not contraindicated
Ethical considerations
The retrospective study received ethical approval from the Human Research Ethical Committee
(HREC) (Appendix B), Site-Specific Assessments (SSA) from the Governance Office of Queensland
Health (Appendix C), Public Health Act (PHA) 2005 (Appendix D) and the University of Queensland
(UQ) Ethical approval (Appendix E). Data were collected from the electronic systems in a de-
identified format with each individual having a unique identifier. The desktop computer used was
password protected and records were stored securely.
Phase 2: VAP education, VAP preventative strategies compliance auditing and surveys
The second phase involved multiple components of research as illustrated in Figure 4.2. However, data
were derived only from auditing compliance of VAP preventative strategies and the perception of the
‘Speaking up for hand hygiene’ initiative via surveys. Two research questions were addressed:
1. What was the compliance with VAP preventative strategies in the QCH PICU between March and
May 2017?
2. What was the perception of the ‘Speaking up for hand hygiene’ initiative amongst parents and
nurses in QCH PICU?
Firstly, in section 4.3.2.1, VAP education for PICU staff and parent were discussed followed by their
engagement with VAP education. Then, in section 4.3.2.2 the VAP preventative strategies
compliance auditing were explained. Finally, surveys on ‘Speaking up for hand hygiene’ initiative
involving parents and PICU nursing staff were discussed in section 4.3.2.3 and 4.3.2.4 respectively.
73
Mar 2017 May 2017
PICU
Staff
Researcher’s strategies on VAP education:
• Approached the PICU educators to reinforce the PICU staff on updated VAP education
package accessible through TEACH-Q platform since August 2016.
Compliance auditing
• Assessed the compliance of seven VAP preventative strategies.
• Provided feedback from compliance auditing of VAP preventative strategies on three
occasions: after 2 weeks of compliance auditing, end of April, and 7 May.
Conducted surveys on ‘Speaking up for hand hygiene’.
Parents
in PICU
Researcher’s strategies on VAP education:
• Provided face-to-face education on hand hygiene and pamphlet, ‘VAP: How I Can Help my
Child in PICU’ was given to parents.
Compliance auditing:
• Assessed the compliance of hand hygiene practises.
Figure 4:2: The research components involved in Phase 2 of the study
VAP education and education engagement
VAP education packages existed at both the Royal Children’s and the Mater Children’s Hospitals prior
to merging as QCH. However, the move, in November 2014, brought about challenges in maintaining
the pre-existing documented practises of VAP preventative strategies as described in Section 4.3.1.6
due to adaptation to the new working environment and systems (Chang, personal communication,
September 2015). The researcher’s role with the updated VAP education package was to update the
existing VAP education based on recent evidence. These findings provided a greater understanding
of the current preventative strategies applied to paediatrics (Cooper & Haut, 2013). Serial meetings
were undertaken with PICU Nurse Educators, and the updated VAP education package was produced
in May 2016. Completion of the VAP education package is compulsory for PICU nursing staff. It is
not compulsory learning for medical practitioners and allied health staff; however, they were
encouraged to enrol.
PICU staff
1. Updated VAP education
The updated education package (Appendix F) was re-launched in August 2016 and involved two
components:
74
1) Ventilator-Associated Pneumonia Prevention: Nursing Interventions with VAP Preventive
strategies.
This is a 32-slide PowerPoint presentation, delivered to PICU staff in the unit via the TEACHQ
platform — an online training system available within the hospital network. The package was the
updated VAP education package, which consists of the content outlined in the table below:
Table 4.2: The content of updated VAP education in PICU
What is VAP and what is the VAP tool?
Outlined in the PNU/VAP tool.
How is VAP diagnosed?
A case presentation of how VAP is diagnosed based on the VAP criteria outlined in the
PNU/VAP.
Focused on the PICU staff’s role in VAP assessment and the VAP surveillance team.
The challenges with the VAP tool
The new ventilator-associated complication tool
An example of study that applied the new ventilator-associated complication to the paediatric population.
Risk factors for VAP
Non-invasive ventilation strategies
Weaning sedation: State Behaviour Scale (SBS)
What is the SBS and the classifications – deep sedation, light sedation and agitation?
Emphasised on the interventions related to SBS scores and the documentation in Metavision
(iMDsoft®).
VAP preventative strategies (bundle)
Emphasised on seven preventative strategies in a poster.
Vigorous hand hygiene
Emphasised on this simple action – remains as primary measure to reduce.
Oral hygiene
Reinforce the mouth care protocol (4-hourly oral hygiene), 12-hourly oral assessment and
documentation.
Cuff pressure checks
PICU guidelines – 12-hourly cuff pressure check, maintain minimum cuff pressure at 10 to 20 cm H2O.
Suction and VAP
Ventilator circuits checks
Head of bed elevation
Early enteral nutrition
Strategies to improve adherence to VAP preventative strategies (bundle)
75
2) VAP Preventive Strategies poster
This is a poster presented on the wall of the patients’ rooms in PICU. The poster was also available
in the PowerPoint slides described earlier. The poster consists of seven VAP preventative strategies
with the images and the expected practises within 24 hours (Table 4.3).
Table 4.3: The content of VAP preventative strategies poster
1. Hand hygiene
Adhere to the 5 Moments for Hand Hygiene.
Always use gloves when: performing suction, disconnecting ETT and breaking ventilator circuit.
2. Oral hygiene
Four-hourly oral hygiene: under six months without teeth – 4-hourly moistening of mouth with sterile
water; above six months – 12-hourly 0.2% Chlorhexidine for patients with teeth and 12-hourly brushing
teeth with toothpaste.
3. Suctioning
ETT – aseptic non-touch technique (ANTT) (open suction)
Oropharynx – sterile bore Y-suction catheter. Use new catheter and suction prior to repositioning;
single-use yankaeur if required.
Equipment – change all suction related equipment 24-hourly.
Saline – syringe of saline single episode use only.
4. Cuff pressure checks
12-hourly cuff pressure checks.
Maintain cuff pressure 10–20cm H2O.
Only deflate cuff fully upon extubation or if excessive swelling occurs.
5. Ventilator circuits
Drain condensate away from patient or expel onto disposable cloth.
Change ventilator circuit at 14 days or if visibly soiled.
Change expiratory filter daily (label with date).
Discard circuit within 24 hours of disconnection and reset ventilator.
6. Head of bed elevation
Keep head of bed 15–30o unless contraindicated.
7. Enteral nutrition
Start enteral nutrition within 24 hours of admission
The expected frequency of performance of each of the VAP preventative strategies in 24 hours was
described in the PICU guidelines. For example, the expected frequency performance of cuff pressure
checks is twice in 24 hours. The majority were stated in the VAP preventative strategies poster and
VAP education in the PowerPoint slides. Otherwise they were mentioned in separate guidelines such
76
as the national standard of hand hygiene compliance set by Hand Hygiene Australia which is more
than 80%, 12-hourly oral health assessment in PICU Oral Hygiene Care Guideline and twice per shift
for ETT suctioning, documented through Metavision (Chang & O'Leary, 2016; Queensland
Children’s Hospital Paediatric Intensive Care Unit, 2015).
2. Provision of feedback from compliance auditing of VAP preventative strategies.
The provision of feedback from compliance auditing of VAP preventative strategies was provided on
three occasions. The first feedback was held at two weeks after commencement (mid-March 2017),
when a presentation was requested in the PICU Patient Safety and Quality Monthly Meeting. The
oral presentation with PowerPoint slides was delivered to PICU staff. The majority of attendees were
nurses. A copy of the presentation was included in the meeting minutes. The second feedback was
delivered through a poster presentation disseminated via the Electronic Information Board. In
addition, daily immediate bedside feedback was provided to PICU staff, particularly for hand hygiene
and for endotracheal suctioning. The final feedback was in the form of a Portable Document Format
(PDF) of slides and circulated by the Nurse Educator to staff email addresses.
Parents
The role of parents in minimising the incidence of VAP is an innovative inclusion to this study.
Educational material for parents was designed that aligned with the Children’s Health Queensland’s
“Speak Up for Safety” initiative. A bi-fold pamphlet, ‘VAP: How I Can Help my Child in PICU’
(Appendix G) was developed and distributed describing simple measures that parents could perform
such as hand hygiene. Parents of children in PICU were provided with face-to-face education with
emphasis on hand hygiene and oral hygiene in PICU.
The development of the pamphlet, ‘VAP: How I Can Help my Child in PICU’, began in mid-April
2016 with a meeting with LCCH PICU clinicians, Lead Nurse of Paediatric Critical Care Research
Group (PCCRG) and a Nurse Educator from the education unit in the PICU. The meeting was a
brainstorming session, sharing information around VAP prevention implementation in the PICU. The
fruitful discussion proposed parental involvement in VAP prevention concerning hand hygiene and
‘Speaking up for hand hygiene’, consistent with the unit interest to empower parents in patient safety.
Subsequently, a series of meetings was undertaken with the Nurse Educator and research supervisors
to finalise a list of VAP preventative strategies which were practical for parents in the PICU, and
education strategies which were suitable for transmitting information. Reliable information was
retrieved from the World Health Organisation (WHO), the Australian Commission on Safety and
Quality in Health Care and QCH PICU Oral Hygiene Guideline and existing VAP education materials
77
available in the unit to draft the content of the pamphlet (Australian Commission on Safety and Quality
in Health Care, 2013; Chang & O'Leary, 2016; World Health Organization (WHO), 2009a). The
content of the pamphlet was revised through five validation rounds involving different panels (refer to
Table 4.4).
Table 4.4: The summary of content validation for bi-fold pamphlet, ‘VAP: How I Can Help my Child
in PICU’
Date Panel members Recommendations/changes
End of May 2016 Two research supervisors, a PICU
clinician and a nurse educator
1. Language should be simple and the
information succinct.
Mid-June Two research supervisors, a PICU
clinician, a nurse educator, a social
worker and two PICU nursing staff
(representing the PICU Safety and
Quality Unit)
1. To change the title of the pamphlet,
simplification and removal of unnecessary
information and images.
July 2016 Two research supervisors, a PICU
clinician, a nurse educator, a social
worker, two PICU nursing staff
(representing the PICU Safety and
Quality Unit) and two parents
1. The language needs to be in line with the
lowest adult health literacy levels
(Australian Bureau of Statistics, 2009).
2. To condense the information to only one
page and add images of hand rub and hand
hygiene using soap.
End of August
2016
Two research supervisors, a PICU
clinician, a nurse educator, a social
worker, two PICU nursing staff
(representing the PICU Safety and
Quality Unit) and two parents
1. Panel approved the pamphlet
2. Final review required from a PICU social
worker.
December 2016 PICU social worker Approved the pamphlet.
February 2017 A PICU clinician and PICU Director 1. Reword the parents’ contribution to care
section “encourages to speak up” to “it is OK
to check if I washed my hands”
March 2017 - Finalised and approval obtained.
Engagement of VAP education
1. PICU staff
A two-fold approach was undertaken for PICU staff education: 1) an update of the VAP education
package and 2) VAP preventative strategies compliance interim feedback (during auditing). These
approaches to VAP education have been previously shown by researchers to be associated with a
reduction of VAP rates (Bigham et al., 2009; Brierley et al., 2012; Flodgren et al., 2013).
The education package was accessed via a compulsory online training platform and education
development program. These were the usual avenues at this hospital and the PICU Nurse Educators
had access to the analytics describing completion rates. The online learning platform is beneficial, as
78
it provides flexibility in access, fitting in with PICU staff needs and shift work (Choules, 2007;
Labeau et al., 2016). Furthermore, staff were provided with a copy of parental pamphlet. This ensured
consistency of information between PICU staff and parents. The majority of staff welcomed this input
and verbalised that they were good reminders for their daily practise.
During compliance auditing, there were also proactive measures from the PICU Patient Safety and
Quality unit. They involved actively promoting ‘Speaking up for hand hygiene’ by performing video
recordings to PICU staff holding a sign, “It is OK to ask me to clean my hands”. The videos were
available around strategic PICU areas (PICU entrances, airlock between two wings of PICUs, pantries
and toilets). Colourful visual reminders such as hands full of germs also posted in those strategic areas
added to the existing posters in addition to monthly hand hygiene compliance conducted by hospital
staff.
Feedback during the compliance auditing occurred via a group presentation, a poster disseminated
via the electronic information board and face to face on the wards. The advantages of presenting at
unit monthly meetings meant that the researcher was able to emphasise the importance of VAP
prevention strategies to various PICU staff and, at the same time, remind them to revisit the education
package on VAP that had been relaunched in August 2016. A soft copy of the presentation was made
available via staff emails as part of their meeting minutes. Staff asked questions concerning their hand
hygiene practise during the presentation, and the researcher provided appropriate answers; mostly
these concerned missed hand hygiene moments. In addition, the researcher also received feedback
from a few staff members who welcomed the auditing and, since the researcher was from the unit it
was easier for them to voice any concerns.
2. Parents
Parents were given the pamphlet, ‘VAP: How I Can Help my Child in PICU’, and face-to-face
education. Parents appeared to be receptive and welcomed the information given. They seemed to
understand the information in the pamphlet and had a few questions regarding the hand hygiene
resources for those who are allergic to the soap or gel provided in the PICU. The VAP education
engagement in PICU is illustrated in Table 4.5.
79
Table 4.5: The VAP education engagement in PICU at Phase 2
Timeline 6 Mar 2017 20 Mar 2017 28 Apr 2017
7 May 2017
Researcher to
PICU staff
1) Reinforcement of updated VAP education by Nurse Educators in TEACHQ platform
Researcher engaged with the PICU staff by:
a) Supplied the bi-fold pamphlet – ‘VAP: How I Can Help my Child in PICU’ designed for parents to align with the Children’s Health
Queensland’s ‘Speak Up for Safety’ initiative through hand hygiene via electronic and printed copies. 2) VAP preventative strategies’ compliance auditing
Researcher engaged with the PICU staff by providing the VAP preventative strategies’ compliance provision of interim feedbacks at three
occasions:
1st VAP compliance auditing’
feedback
via oral presentation in the PICU
Patients Safety Monthly Meeting.
2nd VAP compliance auditing’
feedback via poster presentation
electronic information board and
daily bedside one-to-one
feedback.
3rd VAP compliance auditing’
feedback via PDF slides
circulated to staff email
addresses.
Researcher to
parents
1) VAP education via bi-fold pamphlet and face-to-face education
Patients’ Safety
and Quality Unit
to PICU staff and
parents
1) Promoted ‘Speaking up for hand hygiene’ via video recordings
2) Colourful visual hand hygiene reminders
3) Monthly hand hygiene compliance
80
Compliance auditing
Population and sample
Participants included in hand hygiene compliance auditing were those PICU staff who were caring for
eligible screened patients, including nurses, medical practitioners, allied health clinicians and parents.
Other VAP preventative strategies compliance auditing only involved nurses.
Sample size
Compliance auditing used convenience sampling and aimed to obtain compliance data minimally from
30 to 40 patients based on the sample size calculation. The sample size was determined using EpiInfo
(Centers for Disease Control and Prevention (CDC), 2016a) with the following parameters: 95%
confidence interval, population size, degree of accuracy and 50% of the time the standard will be met
(University Hospitals Bristol NHS Foundation Trust, 2009).
Potential patients were screened for inclusion in auditing prior to recruitment by logging into the
Report Server via QCH Queue Manager and accessing the Patients Summary Report. The patient list
was confirmed with the PICU Nurse Researcher before auditing could commence. This two-stage
screening ensured that only appropriate families were approached for recruitment. For example,
unstable or terminally ill children or those with complex family dynamics were not approached for
recruitment. Patients were followed in PICU from the day of intubation until the day of extubation,
or deceased, or at the audit end date, whichever came first.
Following screening, the applied inclusion criteria were:
• intubated with endotracheal tube (ETT);
• receiving invasive mechanical ventilation.
The exclusion criterion was:
• receiving non-invasive mechanical ventilation.
Audit tools
The VAP preventative strategies checklist from the Dominican Hospital, Santa Cruz, the USA,
published by Healthcare Improvement (IHI) (Institute for Healthcare Improvement (IHI), 2015) was
adapted and a revised version of the VAP audit tool based on the QCH PICU VAP preventative
strategies was produced. The ‘5 Moments for Hand Hygiene’ data collection tool was used to collect
the hand hygiene compliance data (Hand Hygiene Australia, 2017a).
81
The audit tools were piloted in February 2017 to assess practicality and feasibility through two
simulation sessions. The researcher also took note of the common times most of the preventative
strategies were carried out in the unit. The researcher then made changes by adding further strategies:
1) adhered to aseptic non-touch technique (ANTT); 2) adhered to using single suction catheter; and
3) adhered to the use of new normal saline for instillation for every new episode of suctioning.
Data collection
Auditing was carried out in the PICU from the 6th of March until the 6th of May 2017, with
compliance feedback provided in this period (Hand Hygiene Australia, 2017). Two data collection
methods were used: 1) data retrieval from electronic documentation Metavision (iMDsoft®); and 2)
bedside observation. Data for the frequency of oral hygiene performance in 24 hours and the
adherence to the cuff pressure measurements are obtained from Metavision (iMDsoft®), while
adherence to aseptic non-touch technique during ETT suctioning was obtained through bedside
observation (Table 4.6). Auditing was undertaken on four weekdays (Monday to Thursday; excluding
public holidays) morning and/or afternoon (8.00–10.00 am and/or 2.00–4.00 pm).
The researcher determined in consultation with bedside nurses whether any VAP preventative
strategies were due within the next two hours and then returned to conduct the observation. For ETT
suctioning the researcher liaised with the physiotherapist on a daily basis to maximise the opportunity
for suctioning observations. Observations took place either in the patient’s room or just outside the
room. Twenty-minute periods of observation of hand hygiene were carried out to examine compliance
among PICU staff and parents.
Data analysis
Statistical Package for the Social Sciences software IBM version 24 was used to analyse the auditing
data (IBM, 2016). Data were summarised as frequency and percentage, mean and standard deviation,
or median and interquartile range.
Ethical considerations
Permission for auditing was granted by HREC and the Governance of the hospital as part of quality
improvement in the unit. All participating PICU staff and parents were coded with a unique
identifying number.
82
Table 4.6: VAP compliance data auditing
Participants VAP
preventative
strategies
Elements evaluated Measurements Source of data Remarks
Bedside
observation
Metavision
(iMDsoft®)
1. Nurses
2. Medical
practitioner
3. Allied
health
4. Parents
1. Hand hygiene 1. Five moments of hand hygiene
% compliance
(continuous)
Nurses 2. Oral hygiene
(OH)
1. Frequency of performance in 24 hours Continuous
If any chance to observe
the OH performance –
recorded the frequency
of observation
2. Adherence to 12 OH hourly
assessment
Yes/No
3. Age appropriate oral hygiene
cleansing material used.
Yes/No
1. Nurses
2. Allied
health
3. ETT
suctioning
1. Frequency of performance in 24 hours Continuous
Arrangement with
respective staff (bedside
nurse/physio required in
advance e.g. diverse
time of suctioning from
one patient to another)
2. Adherence to aseptic non-touch
technique (ANTT)
Yes/No
Participants VAP
preventative
strategies
Elements evaluated Measurements Source of
data
Remarks
83
Bedside
observation
Metavision
(iMDsoft®)
3. Single suctioning catheter used Yes/No
4. New posiflush saline used every new
episode of suctioning
Yes/No
5. Connected to filtered test lung after
disconnection of ETT and the ventilator
circuit
Yes/No
6. Drain the condensate away from
patient or expel onto disposable cloth
prior to re-connection to ETT
Yes/No
Nurses 4. Head of bed
elevation (HOB)
1. Frequency of performance in 24 hours Continuous
Bedside observation
also performed, and
degree of HOB
compared with the
Metavision (iMDsoft®)
2. Adherence to the degree of HOB Yes/No
Nurses 5. ETT cuff
pressure checks
1. Frequency of performance in 24 hours
(patients who were with cuffed ETT
only)
Continuous
If any chance to observe
the cuff pressure checks
performance – recorded
the frequency of
observation
Participants VAP
preventative
strategies
Elements evaluated Measurements Source of
data
Remarks
Bedside
observation
Metavision
(iMDsoft®)
2. Adherence to 12-hourly cuff pressure
checks
Yes/No
84
3. Maintaining the cuff pressure within
the limit
Yes/No
Nurses 6. Ventilator
circuits checks
1. Frequency of performance in 24 hours Continuous
2. Change expiratory filter every 24
hours
Yes/No
Medical
practitioner
7. Enteral
nutrition
commencement
1. Enteral nutrition commencement
within 24 hours of admission if no
contraindication
Yes/No
85
Survey for parents
The parental survey examined perceptions of VAP prevention following face-to-face education
regarding hand hygiene and oral hygiene in the pamphlet called ‘VAP: How I Can Help my Child in
PICU’.
Population and sample
Parents or primary caregivers of children admitted to PICU during the data collection period were
invited to complete the survey.
Sample size
The total number of participants was based on a sample size calculation performed using the EpiTool
epidemiological calculator (Sergeant, 2017). The latest prevalence of paediatric VAP in Australia
was estimated to be 6.7% (Gautam et al., 2012) and was entered as 0.67 in the calculator as an estimate
of true proportion. The population size of 100 parents was entered in the calculator based on the
estimated number of admissions of children requiring invasive mechanical ventilation to PICU, at
QCH within a three-month period. The Z value was 1.96 and e was 0.05 as desired precision entered
into the calculator respectively. The formula applied was:
𝑛 =Z2 × P(1 – P)
e2
where Z=value from standard normal distribution corresponding to desired confidence level [Z=1.96
for 95% confidence interval (CI)]. P is expected true proportion, e is desired precision (half desired
CI width). The calculated sample size for parents was 78.
The potential parents or primary caregivers for the survey were identified via the Metavision
(iMDsoft®) platform, using a similar screening process to that undertaken for compliance auditing.
Following the screening the applied inclusion criteria were:
• Parent or primary caregiver of a child receiving invasive mechanical ventilation.
• Able to read and understand the English language.
The exclusion criterion was:
• Parent or primary caregiver who did not agreed to participate in the study.
86
Instrument (Survey questionnaire)
The questionnaire used in this study was adapted from publications by Chang, Easterbrook, Hancock,
Johnson, & Davidson, (2010), Kim et al., (2015), Samuel et al., (2012), World Health Organization
(WHO), (2017) and Wu et al., (2013). The questionnaire consisted of three sections: Section A:
Demographic information; Section B: General perception on information provided in the pamphlet:
‘VAP: How I Can Help my Child in PICU’ and Section C: Perceptions of parents about the ‘Speaking
up for hand hygiene’ initiative (Appendix H). Table 4.7 gives a summary of the survey; sections,
number of questions, measurements/scoring and final measurements after the data analysis was
conducted.
The research supervisors evaluated the content validity of the survey questions. The face validity and
overall understanding of the survey questions were tested with five parents in February 2017. Parents
made suggestions and comments, and these were included in the final version of the questionnaire.
These pilot responses were not included in the results.
Data collection
The eligible parents or primary caregivers were identified through a PICU Nurse Researcher before
the education and survey commenced. Following this, one parent or primary caregiver was recruited
to participate in the survey. An explanation regarding the study was given, and face-to-face education
was provided, together with a pamphlet given to the parents. Parents were allowed to ask any
questions and express any concerns. Parents were invited to complete the survey either using the self-
administered questionnaire or an online questionnaire (via Qualtrics™).
Data analysis
The responses from the self-administered questionnaire were manually entered into Statistical
Package for the Social Sciences software (IBM, 2016). Quantitative data were summarised as
frequency and percentage, mean and standard deviation or median and interquartile range. The free
text responses in survey studies were analysed using content analysis (Hsieh & Shannon, 2005). The
responses were highlighted to capture key concepts. Then, these were sorted into categories based on
how different codes were related and linked (Hsieh & Shannon, 2005).
Ethical consideration
Several ethical issues were anticipated. Of most concern was when to approach the parents of
critically ill children. The researcher understood the sensitivity of the situation and the PICU nurse
researchers were approached prior to attending the parents/caregivers or bedside nurses. Additionally,
87
the researcher was educated about referring the parent or primary caregiver to social workers in the
unit if necessary. Informed consent for the survey was important. Survey participants were provided
with explicit information regarding the project, the voluntary obligation of participation, risk and
benefits, confidentiality and the opportunity to express any concerns. This information was included
in the participant information sheet (Appendix I). A response to the online surveys was deemed to
constitute consent of the participant. The survey had approval from the respective ethical bodies
following amendments which included adding the paper version of the survey questionnaire for the
parents.
88
Table 4.7: Elements in parents’ survey
Section Number of questions/statements
Measurement/scoring Final measurement
A: Demographic
information
Six:
• Age
• Gender
• Education
• Working in healthcare field or not
• History of child admission to PICU
• History of their child who received mechanical
ventilation via ventilator in PICU
• Tick in the designated boxes
(Categorical)
• Age was collapsed into 2
categories from 3
Less than 30 years old and more
than 31 years old
• Education level was collapsed
into 2 categories from 5
Non-formal qualification
Formal qualification
B: General
perception on
information
provided in the
pamphlet: ‘VAP:
How I Can Help
my Child in
PICU.”
Ten:
• One question about whether the parent ever
heard about VAP before
• One question about the method they routinely
use for hand hygiene in PICU
• Eight statements referred to the level of
agreement with the information stated in the
pamphlet
• Tick in the designated boxes
Categorical: yes, no, unsure
• Hand gel, hand wash,
combination
• 5-point Likert Scale
[1=strongly disagree,
5=strongly agree]
• 5-point Likert Scale was
collapsed into disagree (1, 2, 3)
and agree (4, 5).
C: General
perception on
information
provided in the
pamphlet:
‘Speaking up for
hand hygiene’
Nine:
• Six statements referred to the level of
agreement with the information stated in the
pamphlet
• Two questions on the possible reason parents
would stop to remind nurses and other PICU
staff if the parents saw that they did not
perform hand hygiene
• One open-ended question for suggestions or
comments to improve the hand hygiene
practise in the unit.
• 5-point Likert Scale
[1=strongly disagree,
5=strongly agree]
• Tick in the designated boxes
Categorical: four options and
other reason with free text
response
• Free text response
• 5-point Likert Scale was
collapsed into disagree (1, 2, 3)
and agree (4, 5).
89
Surveys for nurses
Nursing staff within the unit were surveyed to establish the priority they placed on their perception of
the ‘Speaking up for hand hygiene’ initiative.
Population and sample
All PICU nurses were invited to participate in the study.
Sample size
The total number of participants was based on a sample size calculation using the EpiTool
epidemiological calculator (Sergeant, 2017). The prevalence of paediatric VAP in Australia was
estimated to be 6.7% (Gautam et al., 2012) and this was entered as 0.67 in the calculator as estimated
true proportion. The population size of 150 nurses was entered in the calculator based on the estimated
number of nurses employed in PICU. Z value was 1.96 and e was 0.05 as desired precision entered
into the calculator respectively. The formula applied was:
𝑛 =Z2 × P(1 – P)
e2
where Z=value from standard normal distribution corresponding to desired confidence level [Z=1.96
for 95% conficence interval (CI)]. P is expected true proportion, e is desired precision (half desired
CI width). The calculated sample size for nurses was 105.
The inclusion and exclusion criteria
The inclusion criterion was:
• PICU nursing staff.
The exclusion criterion was:
• Nurses who have not agreed to participate in the study.
Instrument (Surveys questionnaire)
The questionnaire used for data collection was adapted from Kim et al., (2015), Samuel et al., (2012),
World Health Organization (WHO), (2017) and Wu et al., (2013). The questionnaire consisted of two
sections: Section A: Demographic information, and Section B: Perception of nurses on the ‘Speaking
up for hand hygiene’ initiative (Appendix J). Table 4.8 gives a summary of the sections of the
questionnaire, number of questions, measurements, and final measurements after the data analysis
90
was conducted. The questionnaire was initially available online (via Qualtrics™) and subsequently
made available in hard-copy form. The online questionnaire was distributed via a link to PICU
nursing staff members’ email addresses.
Data collection
The online survey invitation via the Qualtrics™ platform was sent to all nurses through the PICU
Nurse Researcher as advised by the Lead Nurse of PCCRG (Qualtrics, 2016). The first invitation was
sent on the 1st of April 2017, and the researcher gave verbal reminders to the nurses while conducting
the audit. In early May 2017, a reminder email was sent by the Lead Nurse of PCCRG. A hard-copy
version was also distributed to increase the response rate. The survey period was from the 1st of April
to the 6th of June 2017. The researcher went to each PICU nursing staff and asked did he/she had
attempted to participate the online version before. Those who did not participate in online version, an
explanation regarding the study was given. The nurses’ information sheet and hard copies of survey
were given. Hard copies were then collected from the designated survey boxes in PICU, every day in
the afternoon on the following day.
Data analysis
All online responses were recorded in Qualtrics™ (Qualtrics, 2016) and the self-administered
questionnaires were manually entered into Statistical Package for the Social Sciences software (IBM,
2016). The free text responses in survey studies were analysed using content analysis where the
responses were highlighted and categorised (Hsieh & Shannon, 2005). Quantitative data were
summarised as frequency and percentage, mean and standard deviation or median and interquartile
range.
Ethical considerations
Nurses were also provided with the nurses’ information sheet which provided explicit information
regarding the study, the voluntary obligations of participation, risks and benefits, confidentiality and
the opportunity to express any concerns. A response to the online survey constituted informed consent
by the participant.
91
Table 4.8: Elements in nurses’ survey
Section Number of questions/statements
Measures Final measurement
A: Demographic
information
Three
• Gender
• Length of working experience in PICU
• The method nurses routinely use for hand
hygiene in PICU
• Tick in the designated boxes
(Categorical)
• Free text response
• Categorical: yes, no, unsure;
hand gel, hand wash,
combination
• Hand hygiene method nurses
used was collapsed from 3
categories into 2 categories; a
single method only or
combination.
B: General
perception on
‘Speaking up for
hand hygiene’
Nine
• Six statements referred to the level of
agreement of ‘Speaking up for hand hygiene’
• Two questions on the possible reason nurses
would stop to remind parents and other PICU
staff if the nurses saw they did not perform hand
hygiene
• One open-ended question for suggestions or
comments to improve the hand hygiene practise
in the unit.
• 5-point Likert Scale
[1=strongly disagree,
5=strongly agree]
• Tick in the designated boxes
Categorical: three options and
other reasons with free text
response
• Free text response
• 5-point Likert Scale was
collapsed into disagree (1,2,3)
and agree (4,5)
92
Phase 3: Prospective study
In this final phase, the incidence of VAP and VAE was examined. Four research questions were
addressed:
1. What was the incidence of VAP and VAE in QCH PICU from the 12th of June until the 12th
of December 2017?
2. What was the compliance with VAP preventative strategies in QCH PICU from the 12th of
June until the 12th of December 2017?
3. Did VAP education and compliance preventative strategies including auditing with interim
feedback reduce VAP and VAE status as defined by the PNU1/VAP surveillance and the VAE
surveillance tool?
4. Was there any improvement in VAP preventative strategies compliance after the VAP
education and compliance preventative strategies auditing feedback?
Population and sample
All children admitted to QCH PICU requiring invasive mechanical ventilation for greater than or
equal to 48 hours during the period from 12 June 2017 to 12 December 2017 were screened. The
prospective surveillance involved all admissions to QCH PICU during this time period.
The inclusion criteria were:
• age 0 to 18 years old
• invasively mechanically ventilated for ≥ 48 hours
• intubated with an ETT.
The exclusion criteria were:
• non-invasive ventilation
• ventilated for < 48 hours
• admission with existing tracheostomy intubation.
Data collection
The recruitment was based on intubation admission reports that were screened daily to determine
eligible patients. The reports were available within Children’s Health Queensland servers at Children
Health Research Centre. Parents of eligible patients were approached for informed consent. The
eligible patients were followed over time from intubation until the patients were extubated, died, or
93
data collection ceased. Daily data collection ended on the 12th of December 2017 at 00:00 hours. The
data collection via the electronic repositories mentioned in 4.3.1.2 were again utilised and entered
into a Microsoft Excel spreadsheet.
VAP and VAE surveillance tools
The study utilised the criteria from the PNU1/VAP and the VAE surveillance tools described in
4.3.1.3.
The demographic characteristics, possible risk factors, and preventative strategies mirrored to the
retrospective study were also included in this phase. VAE risk factors such as steroid and blood product
administration were added in this phase based on recently published literature (Cocoros, Priebe, Gray
et al., 2017) (Appendix K).
Data analysis
Data were analysed using the Statistical Package for the Social Sciences software IBM version 24,
(IBM, 2016). Process for data checking, exporting, analyses and presentation were similar to the
process used in the retrospective study previously described in section 4.3.1.7. In the prospective
study, the incidence of VAP and VAE and the compliance of VAP preventative strategies were
additionally compared with the retrospective study findings, but only at univariate analysis level.
Ethical considerations
Parents were provided with general information about the research topic, the research objectives,
duration of the study and the instruments used to collect the data. Parents were also informed about
potential risks and benefits, confidentiality, voluntary basis of participation and the opportunity to
express any concerns related to the study. This information was provided in the Parents/Guardians
Information Sheet and Consent Form (Appendix L). The amendment for waiver of consent was
considered since there were a number of admissions with a high probability of being identified with
VAP and VAE, but the child was very ill or deceased before the researcher was able to obtain consent.
In each of these cases it would have been inappropriate to approach parents for consent. Considering
the value of this data the discussion held with the PICU research group and advisory team the potential
to apply for a waiver of consent. Agreement was reached, and an application was sent to respective
ethical bodies. In mid-October 2017, approval for a waiver of consent by HREC was granted
(Appendix M).
94
4.4 Summary
This chapter outlines the methodological approaches used to provide a thorough investigation of VAP
and VAE in critically ill children. Phase 1 provided a picture of VAP/VAE prevalence at baseline.
Phase 2 explored VAP compliance auditing and surveys. In this second phase, various components
were involved: 1) reinforcement of the VAP education package for PICU staff and education for
parents regarding hand hygiene; 2) VAP preventative strategies compliance auditing with feedback,
involving PICU staff, i.e., nurses, medical practitioners, allied health staff, and parents. Surveys to
find perceptions of ‘Speaking up for hand hygiene’ amongst parents and nurses were also undertaken.
Finally, Phase 3 involved a prospective study to assess the effects of the intervention.
The next chapter will highlight the results and discussion for Phase 1: Retrospective study. Chapter
6 presents the results and discussion for Phase 2: VAP preventative strategy compliance auditing and
surveys. Chapter 7 presents the results and discussion for Phase 3: Prospective study.
95
Results and discussion Phase 1: Retrospective study
Introduction
This chapter describes the results and discussion for Phase 1: Retrospective study. The aim of the
retrospective study was to describe the baseline of VAP/VAE incidence and prevalence in the PICU
of the Queensland Children’s Hospital (QCH). This chapter addresses the following research
questions:
1. What was the incidence and prevalence of VAP and VAE in QCH PICU in 2015?
2. What is the sensitivity and specificity of the VAE surveillance tool compared to the PNU1/
VAP surveillance tool?
3. What was the compliance of VAP preventative strategies in QCH PICU in 2015?
4. What were the possible risk factors and preventative strategies associated with the diagnosis
of VAP and VAE in QCH PICU in 2015?
The methodology for this chapter was described in Chapter 4, Section 4.3.1.
Selection of patients with eligible mechanical ventilation episodes
Of the 669 episodes of mechanical ventilation reviewed in the PICU of the QCH, 262 met the required
inclusion and exclusion criteria (Figure 5.1) with an inclusive total of 1713 mechanical ventilation
days, median duration of four days. The PNU1/VAP and the VAE surveillance tool was thus applied
to 262 episodes of mechanical ventilation. For VAP and VAE, the total time until patients were no
longer at risk (i.e., ventilation ceased) or, they were diagnosed with VAP or VAE as per the tool, or
they died, was 1562.27 and 1532.48 mechanical ventilation days respectively.
96
Figure 5:1: 262 episodes of mechanical ventilation met the study criteria
Demographic characteristics of patients in PICU admission
Of the 253 admissions in this study, there were 234 eligible patients. Males accounted for 57.3% of
the eligible patients. More than half the patients were aged less than 1 year (62.5%). The majority
(64.4%) of PICU admissions in 2015 that required mechanical ventilation ≥48 hours were for a
medically-associated condition; 37.0% for an underlying cardiovascular problem and 43.1% from a
direct admission. The median Paediatric Index of Mortality Score 3 (PIMS3) was 1.90 (IQR: 0.70 –
5.50). These indicate that the overall probability of death in this cohort was 1.90% (Table 5.1).
Excluded 22 that receiving invasive
ventilation via tracheostomy
10 mechanical ventilation episodes were
merged into 1, 2, 3 or 4 mechanical
ventilation episodes after re-evaluation
for any breaks between datasets that the
mechanical ventilation was < 24hrs
262 mechanical ventilation
episodes
(234 patients, 253 admissions;
1713 total mechanical ventilation
days)
294 had invasive ventilation
episodes for ≥48 hours
(245 patients)
272 receiving invasive ventilation
via ETT
(234 patients)
669 patients with invasive
ventilation and intubation episodes
in 2015
(252 patients)
Excluded 375 which had intubation
episodes for <48hrs
97
Table 5.1: Demographic characteristics according to patient admission (n=253)
Characteristics n (%)
Gender Male 145 (57.3)
Female 108 (42.7)
Age <1 year 158 (62.5)
1- 12 year 80 (31.6)
13 and above 15 (5.9)
Weight (kg, median (IQR)) 5.8 (3.6- 15.0)
PICU source of admission
OT/ Recovery 70 (27.7)
Emergency
Department
37 (14.6)
Ward (other
inpatient area)
37 (14.6)
Direct admission 109 (43.1)
PICU diagnosis category
Medical 163 (64.4)
Surgical 71 (28.1)
Trauma 19 (7.5)
Underlying disease
Trauma/ Injury 29 (11.5)
Cardiovascular 94 (37.2)
Neurological 21 (8.3)
Respiratory 50 (19.8)
Renal 3 (1.2)
Gastrointestinal 10 (4.0)
Infection 17 (6.7)
Miscellaneous 29 (11.5)
PIMS3 (median, IQR)
PIMS3 1.90 (0.70- 5.50)
OT=Operation Theatre; PIMS3=Paediatric Index of Mortality Score 3; IQR=Interquartile range
Outcome variables according to mechanical ventilation episodes
Table 5.2 shows the outcome variables according to mechanical ventilation episodes. Patients had a
median duration of mechanical ventilation of 4.2 days (IQR: 2.8- 7.5) and stayed in PICU for 7.6
days (median) and 18.9 days (median) in the hospital respectively. Of 234 patients, 85.5% were
discharged to the ward and then home. In the 2015 cohort, 25 patients died.
98
Table 5.2: Outcome variables according to mechanical ventilation episodes (n=262)
Variables n (%)
Duration of mechanical
ventilation
(days, median (IQR))
4.2 (2.8- 7.5)
Length of PICU stay
(days, median (IQR))
7.6 (4.8- 13.5)
Length of hospital stay
(days, median (IQR))
18.9 (9.8- 34.4)
PICU outcome
Discharge to ward/home 227 (86.6)
Died 26 (9.9)
Transferred to another ICU (includes Neonatal
ICU)
9 (3.4)
Mortality (*n= 234)
Yes 25 (10.7)
No 209 (89.3)
* patient level
VAP and VAE counts according to mechanical ventilation episodes
Of 262 mechanical ventilation episodes, 16 (6.1%) were identified as having VAP using the
PNU1/VAP surveillance tool and 16 (6.1%) were identified as have VAE using the VAE surveillance
tool. All 16 met the VAC tier; four of these met the infection-related ventilator-associated
complication (IVAC) tier, and three met the possible ventilator-associated pneumonia (PVAP) tier.
The majority (14 out of 16) of patients identified as having VAP. Ten out of 16 patients identified
with VAE. All happened during their single admission to PICU with a single mechanical ventilation
episode. Four patients with multiple PICU admissions were identified with VAP or VAE at their first
PICU admission, one for VAP and three for VAE. One patient with a single admission was identified
with both VAP and VAE during the second mechanical ventilation out of three mechanical ventilation
episodes in total.
Incidence and prevalence of VAP in PICU
The incidence rate of VAP was 9.3 per 1000 ventilation days according to the CDC tool (end point
being end of ventilation as the denominator) and 10.2 per 1000 ventilator days until the patients were
no longer at risk of VAP (that is ventilation ceased, they were diagnosed with VAP as per the tool, or
died). There were 16 episodes of mechanical ventilation from 16 patients (6.8%) identified as VAP
using the PNU1/VAP surveillance tool, giving a prevalence of 6.1%. No patients were identified as
having VAP on more than one occasion.
99
Incidence and prevalence of VAE in PICU
The incidence rate of VAE was 9.3 per 1000 ventilator days according to the CDC tool (end point
being end of ventilation as the denominator) and 10.4 per 1000 ventilator days defined by end points
at which the patients are no longer at risk of VAE (that is ventilation ceased, they were diagnosed
with VAE as per the tool, or died). There were 16 mechanical ventilation episodes met the VAE
surveillance tool, a prevalence of 6.1%. Fifteen patients (6.4%) were identified as having VAE with
one patient experiencing two VAC/VAEs on two separate occasions.
The sensitivity and the specificity of the VAE surveillance tool
The sensitivity of the new VAE surveillance tool was 18.8% (95% CI: 4.1–45.7) and its specificity
was 94.7% (95% CI: 91.3–97.2). This indicates that the VAE surveillance tool is a poor surveillance
tool for correctly diagnosing VAP as it has a high number of false negative results. Conversely it is a
very good surveillance tool for correctly identifying patients who test negative for VAP as it has few
false positive results.
The positive predictive value (PPV) was 18.8 (95% CI: 6.8–42.1) and negative predictive (NPV)
value 94.7 (95% CI: 85.8–95.8), indicating that a patient is highly likely not to have the disease given
that the test result is negative. The low PPV reflects the low prevalence of VAP. Hence, the VAE
surveillance tool does not appear to be a useful surveillance tool for detecting VAP, but it is useful
for screening patients who do not have the disease.
A patient who had VAE and VAP in the same mechanical ventilation episode was further classified
as IVAC (using the VAE surveillance tool) while identified as VAP using the PNU1/VAP
surveillance tool. Agreement in the identification of VAP and VAE occurred during three episodes
of mechanical ventilations and the surveillance tools agreed VAP and VAE was not present during
233 episodes of mechanical ventilations (Table 5.3).
Table 5.3: Contingency table of VAE versus VAP
VAP
No Yes Total
VAE No 233 13 246
Yes 13 3 16
Total 246 16 262
100
Agreement between the two surveillance tools
Cohen’s κ was run to determine if there was agreement between the two methods in detecting
pneumonia. There was slight agreement between the methods, κ=0.135 (95% CI, 0.055 - 0.325), p=
0.029.
Discussion
Discussion of demographic characteristics of the patients
In this retrospective study, eligible episodes of invasive mechanical ventilation were assessed based
on criteria that have been widely used by previous VAP studies (Elward et al., 2002; Gautam et al.,
2012). This is important to distinguish between healthcare-associated infections (HAIs) and
community-acquired infections. The cut-off point of ≥48 hours was used in line with previous studies
for VAP (Charles et al., 2014; Elward et al., 2002; Kusahara et al., 2014; Roeleveld et al., 2011).
In the present study, the number of patients who were mechanically ventilated for ≥ 48hours in a 35-
bed PICU (234 patients) was lower than the number in a retrospective study of a developed country
(Japan) where 320 patients were observed over one- year period in an eight-bed PICU (Hatachi et al.,
2015). Most of the patients in this study were aged less than 12 months and they were predominantly
male (57.3%), which was consistent with the report of the Australian and New Zealand Paediatric
Intensive Care Registry in 2015 (Australian and New Zealand Intensive Care Society, 2016). The
annual report states that infants aged <12 months constitute 56.1% of admission, with 66.4% aged
less than five years. According to the same report, the overall admission rate to PICU for males was
2.15 per 1000 children compared to females at 1.67 per 1000 children (Australian and New Zealand
Intensive Care Society, 2016). Several studies demonstrate that younger children have a higher risk
for device-associated infections such as VAP (Becerra et al., 2010; Grohskopf et al., 2002; Tang et
al., 2009).
In contrast to the annual report of the Australian and New Zealand Paediatric Intensive Care Registry
in 2015, an underlying disease due to respiratory causes accounted for half (51.5%) of PICU
admissions (Australian and New Zealand Intensive Care Society, 2016). The majority of patients
(64.4%) in this study had an underlying disease due to cardiovascular causes. This may influence the
variability of oxygenation requirement for VAP/VAE diagnosis, for example with congenital
cyanotic heart disease and veno-arterial shunting (Cocoros et al., 2016).
The median PIMS3 value was 1.9%, which is lower than the reported mean for children in Australasia
of 2.9%. In the United Kingdom and Ireland the mean PIMS3 value was at 3.9% (Straney et al.,
101
2013). This difference may be due to the admission threshold, baseline population health status, or
individual patient management (Straney et al., 2013).
Discussion of VAP and VAE in the PICU of the QCH in 2015
The VAP incidence rate of 9.3 per 1000 ventilator days found in this study was within the expected
range for developed countries of 1.8–17.1 per 1000 ventilator days (Hatachi et al., 2015; Ismail et al.,
2012; Muszynski et al., 2013; Roeleveld et al., 2011; Stabouli et al., 2012).
In contrast, the incidence rate of VAP in the present study was higher in comparison to the latest
single setting one-year prospective study in Australia with 7.02 per 1000 ventilator days and a
retrospective study by Iosifidis et al. (2015) which showed 7.7 per 1000 ventilator days. It is, however,
difficult to directly compare the incidence rate to other paediatric studies, due to the diversity of
paediatric demographics, institutional preventative strategies and treatment options (Gautam et al.,
2012; Nair & Niederman, 2015). The incidence rate found in the present study may be due to the
challenges in implementation and monitoring of VAP preventative strategies when the QCH was
moved to a new site in November 2014 and the rotation of new staff/senior doctors, nurses and other
PICU staff that also occurred at that time (Siegel, Rhinehart, Jackson, Chiarello, & Health Care
Infection Control Practises Advisory, 2007).
The VAE incidence rate in this study (9.3 per 1000 ventilator days) is lower compared to two previous
studies which reported 11.2 and 20.9 per 1000 ventilator days respectively (Iosifidis et al., 2016;
Phongjitsiri et al., 2015). However, this incidence rate was higher than three studies that reported
incidence rates ranging from 1.1–4.6 per 1000 (Beardsley et al., 2016; Cocoros et al., 2016;
Narayanan et al., 2016). The higher VAE/VAC identified in this current study may partly be due to
the dominant proportion of patients with underlying cardiac disease, as suggested by Cocoros et al.
(2016), who tested FiO2 threshold and found a higher VAC incidence in their CICU patients. Other
possible explanations may be due to the large diversity of patient characteristics and underlying
diseases included in those studies. For example, Cocoros et al. (2016) included CICU, Neonatal ICU
and PICU patients. Although Narayanan et al. (2016) and Beardsley et al. (2016), conducted their
study in a PICU, one was prospectively conducted for a six-month period and Beardsley et al. (2016),
potentially underestimated the true incidence of VAE by only reporting IVAC. Furthermore, there
were also modifications of the VAE surveillance tool by Cocoros et al. (2016) — they replaced PEEP
with mean airway pressure (MAP) at different thresholds for VAC. In addition, Beardsley et al.
(2016) applied daily minimum PEEP values of at least 2cmH2O instead of 3cmH2O, which may have
contributed to the lower incidence of VAE.
102
The present study revealed equal numbers of episodes of VAE and VAP detected by two surveillance
tools for a total of 16 episodes (in 15 patients for VAE and 16 patients for VAP). This has not been
reported by previous studies. Nevertheless, this result mirrored the findings of a single setting PICU
study by Iosifidis et al. (2016), which identified 12 patients identified as VAE versus 13 patients
identified as VAP.
Other previous studies have reported higher VAE and VAP counts using two surveillance tools (the
VAE (adult) versus PNU/VAP); 41 patients versus nine patients (Phongjitsiri et al., 2015); seven
patients vs. four patients (Narayanan et al., 2016); and 17 patients vs. 15 patients (Taylor et al., 2014).
Conversely, one study reported lower VAE counts as compared to VAP using two surveillance tools
(the VAE versus PNU/VAP); four mechanical ventilation episodes versus five mechanical ventilator
episodes (Beardsley et al., 2016).
This present study found that only four patients met both surveillance tools, with only three
mechanical episodes that matched PNU1/VAP surveillance tool and PVAP tier (VAE surveillance
tool) and this indicates poor agreement between the two surveillance tools. This result was similar to
the previous study by Iosifidis et al. (2016), where five patients out of 25 patients met both
surveillance tools. Only one mechanical ventilation episode out of nine mechanical ventilation
episodes met both surveillance tools in the Beardsley et al. (2016) study, although their study differed
with the application of modified PEEP values of 2cmH2O. Cocoros et al. (2016) also found low
concordance of the VAE surveillance tool in comparison to the PNU/VAP tool. These low agreements
have also been noted in adult studies (Klein Klouwenberg et al., 2013; Stevens et al., 2014).
The sensitivity and specificity result for the VAE surveillance revealed that it is useful for screening
paediatric patients who do not have VAP (high specificity), which may reflect the ability of this tool
to capture other ventilator-associated complications (Hayashi et al., 2013; Klompas, Kleinman, et al.,
2014). Using VAE and surveillance tools in the present study on a single mechanical ventilation
episode in one patient, this tool was able to discriminate the episode as being either IVAC or PVAP,
while using PNU1/VAP, this was classified as VAP. Similarly, in an adult study, using the VAE
surveillance tool, fewer cases of VAP (PVAP) were identified. This reflected the ordinal tiers (VAC
and IVAC) applicable in VAE surveillance tool was able to distinguish VAP (PVAP) (Boyer et al.,
2015). Thus, the introduction of a VAE surveillance tool, with the purpose to capture not only VAP
but also other ventilator-associated complications, may echo those of others who have reported
limitations to the present PNU/VAP surveillance tool in children (Klompas, 2013a; Septimus et al.,
2015).
103
Potential risk factors for VAP and VAE
Table 5.4 describes the potential risk factors encountered in the 262 mechanical ventilation episodes
studied. There were fewer reintubations during the course of mechanical ventilation in the 2015
cohort (74.0%). The majority of patients (78.6%) received no paralytic agent and were less sedated
(94.3%). Over half (53.4%) of mechanical ventilation episodes did not have gastrointestinal
prophylaxis (GI) prescribed. Nasal endotracheal tubes were more common in mechanical ventilation
episodes (64.5%).
Table 5.4: Risk factors according to mechanical ventilation episodes (n=262)
Risk factors n (%)
Reintubation
Yes 68 (26.0)
No 194 (74.0)
Paralytic agent
Yes 56 (21.4)
No 206 (78.6)
Gastrointestinal prophylaxis (GI)
Yes 122 (46.6)
No 140 (53.4)
Nasogastric presence
Yes 259 (98.9)
No 3 (1.1)
Routes of intubation
Nasal 169 (64.5)
Oral 93 (35.5)
Sedation level
Deep sedation 15 (5.7)
Light sedation 247 (94.3)
Compliance of preventative strategies for VAP
Table 5.5 describes the compliance of VAP preventative strategies implemented in 2015. Two VAP
individual preventative strategies which had achieved full compliance were hand hygiene and
performance of endotracheal tube (ETT) suctioning in accordance with the Australian National
Standard and PICU Standard. Oral hygiene was performed regularly (4.9 (SD: 1.0)) with an overall
percentage compliance of 81.6%; however, the frequency of performance did not meet the unit’s Oral
Hygiene Guideline (6 times/day). The head of bed (HOB) elevation and cuff pressure checks were
reported to have a 90.0% compliance rate. The lowest compliance was with ventilator circuit checks
at 70.8%. The overall compliance of VAP preventative strategies (VAP bundle) in the PICU of the
QCH in 2015 was 89.0%.
104
Table 5.5: Comparison between VAP preventative strategies compliance and PICU/national standard
practise
VAP Preventative Strategies Mean (SD)
PICU/ national
standard
% Compliance in
comparison to PICU
/national standard
Hand hygiene 86.4 (12.3) > 80% ≥100%
Oral hygiene 4.9 (1.0) 6 times/24hours 81.6
ETT suctioning 8.6 (2.6) 4 times/24hours ≥100%
HOB elevation (median (IQR)) 22 (19.4-23.4) 24 times/24hours 91.6
Cuff pressure checks
(median (IQR))
1.8 (0.9- 3.0) 2 times/24 hours 90.0
Ventilator circuits checks 17 (6.6) 24 times/ 24
hours
70.8
SD=Standard deviation; IQR=Interquartile range
Univariate analysis
The univariate analysis was conducted to assess for associations between possible explanatory
variables and the outcome of either VAP or VAE in the study as the primary outcome.
Association between demographic characteristics and VAP and VAE
Univariate analysis revealed no association between demographic characteristics and development of
VAP/VAE. Underlying respiratory disease was associated with the development of VAE in all
mechanical ventilation episodes (p=0.045) (Table 5.6). Children with either underlying medical or
surgical PICU diagnosis categories were associated with the development of VAP, but this did not
reach statistical significance (p= 0.062).
105
Table 5.6: Univariate analysis for association of demographic characteristics with VAP/VAE (n=262)
Variables VAP p-
value
VAE p-
value
Yes No Yes No
Gender Male 7 (43.8%) 139 (56.5%) 0.44 9 (56.3%) 137 (55.7%) 0.97
Female 9 (56.3%) 107 (43.5%) 7 (43.8%) 109 (44.3%)
Age <1 year 8 (50.0%) 159 (64.4%) 0.34 12 (75%) 155 (63.0%) 0.57
1-12 year 6 (37.5%) 74 (30.1%) 3 (18.8%) 77 (31.3%)
13 and above 2 (12.5%) 13 (5.3%) 1 (6.3%) 14 (5.7%)
Weight 9 (4.7-28.8) 5.5 (3.5-15.0) 0.17 4.8 (3.3-
15.4)
5.8 (3.6- 15.0) 0.71
PICU
Diagnosis
Category
Medical 14 (87.5%) 158 (64.2%) 0.062 12 (75.0%) 160 (65.0%) 0.59
Surgical 2 (12.5%) 88 (35.8%) 4 (25.0%) 86 (35.0%)
PICU
source of
admission
OT/
Recovery
3 (18.8%) 67 (27.2%) 0.64 4(25.0%) 66 (26.8%) 0.22
Emergency
Department
3 (18.8%) 35 (14.2%) 3(18.8%) 35 (14.2%)
Ward (other
inpatient
area)
4 (25.0%) 37 (15.0%) 5(31.3%) 36 (14.6%)
Direct
admission
6 (37.5%) 10.7 (43.5%) 4(25.0%) 109 (44.3%)
Under-
lying
disease
Cardio-
vascular
4 (25.0%) 93 (37.8%) 0.33 10(62.5%) 87 (35.4%) 0.045
Respiratory 2 (12.5%) 48 (19.5%) 0 (0.0%) 50 (20.3%)
Others 10 (62.5%) 105 (42.7%) 6 (37.5%) 109 (44.3%
PIMS3
PIMS3 1.80 (1.30-
6.00)
2.00 (0.70-
6.30)
0.99 2.70 (1.60-
7.80)
1.80 (0.70-
5.30)
0.091
PIMS3= Paediatric Index of Mortality Score 3; IQR= Interquartile range
Association of outcome characteristics with VAP and VAE
There was an association between duration of mechanical ventilation, length of PICU and hospital
stay and VAP and VAE occurrence (p<0.05) (Table 5.7). No association was found between PICU
outcome, mortality, VAP or VAE.
106
Table 5.7: Univariate analysis for association of patient outcome characteristics with VAP and VAE
(n=262)
Association of possible risk factors with VAP and VAE
Reintubation and the presence of GI prophylaxis were associated with the development of VAE (p <
0.05, Table 5.8). Although not statistically significant, there was some evidence to suggest that both
are associated with the development of VAP (p=0.14 and p=0.07). The use of a paralytic agent, route
of intubation and sedation level was not found to be associated with either VAP or VAE.
Variables VAP p-
value
VAE p-
value
Yes No Yes No
PICU
outcomes
Discharge
ward/home &
Trans to
another
ICU/Neonatal
ICU
13 (81.3%) 223 (90.7%) 0.20 13 (81.3%) 223 (90.7%) 0.38
Died in ICU 3 (18.8%) 23 (9.3%) 3 (18.8%) 23 (9.3%)
Mortality
Died
3 (18.8%)
23 (9.3%)
0.20
3 (18.8%)
23 (9.3%)
0.20
Not died 13 (81.3%) 223 (90.7%) 13 (81.3%) 223 (90.7%)
Duration of
mechanical
ventilation
(days, median
(IQR))
11 (5.2-19.6) 4.1 (2.8-6.8) 0.001 13.7 (8.2- 20.5) 4.1(2.8-6.5) 0.001
Length of
PICU stay
(days median
(IQR))
17.8 (10.9-
34.8)
7.2 (4.7-12.8) 0.001 32.8 (26.6- 52.5) 6.9 (4.7- 12.3) 0.001
Length of
hospital stay
(days median
(IQR))
33.1 (17.7-
85.2)
18.7 (9.3- 33.6) 0.007 51.2 (29.7- 64.5) 9.3 (18.1-32.6) 0.001
107
Table 5.8: Univariate analysis for association of possible risk factors with VAP and VAE (n=262)
Variables VAP p-
value
VAE p-
value
Yes No Yes No
Reintubation Yes 7 (43.8%) 61 (24.8%) 0.14 10 (14.7%) 6 (3.1%) 0.002
No 9 (56.3%) 185 (75.2%) 58 (14.7) 188 (96.9%)
Paralytic agent Yes 4 (25.0%) 52 (21.1%) 0.75 4 (7.1%) 52 (92.7%) 0.75
No 12 (75.0%) 194 (78.9%) 12 (5.8%) 194 (94.2%)
GI prophylaxis Yes 11 (68.8%) 111(45.1%) 0.07 14 (11.5%) 108(88.5%) 0.001
No 5 (31.3%) 135 54.9%) 2 (4.1%) 138 (98.65)
Nasogastric
presence
Yes 16 (100%) 243 (98.8%) 1.00 16 (6.2%) 243 (93.8%) 1.00
No 0 (0%) 3 (13.0%) 0 (0%) 3 (100%)
Routes of
intubation
Nasal 9 (56.3%) 160 (60.0%) 0.48 12 (7.1%) 157 (92.9%) 0.27
Oral 7 (43.8%) 86 (35.0%) 4 (4.3%) 89 (95.7%)
Sedation level Deep
sedation
1 (6.3%) 14 (5.7%) 1.00 0 (0%) 15 (100%) 0.61
Light
sedation
15 (93.8%) 232 (94.3%) 16 (6.5%) 231(93.1%)
Association of preventative strategies with VAP and VAE
Univariate analysis revealed that cuff pressure checks were associated with VAP and VAE occurrence
(p ≤ 0.05). Although no significant association was found, there is some evidence to suggest that oral
hygiene is associated with the development of VAE (p= 0.092) (Table 5.9).
108
Table 5.9: Univariate analysis for association of VAP preventative strategies with VAP/VAE (n=262)
Variables VAP p-
value
VAE p-
value
Yes No Yes No
Hand hygiene
(mean (SD))
88.8 (14.7) 86.2 (12.1) 0.44 85.5 (13.9) 86.5 (12.2) 0.75
Oral hygiene
(mean (SD))
4.8 (0.7) 4.9 (1.0) 0.75 5.6 (1.1) 4.9 (1.0) 0.092
ETT suctioning
(mean (SD))
8.7 (2.2) 8.5 (2.7) 0.86 8.1 (1.7) 8.6 (2.7) 0.45
HOB elevation
(median (IQR))
22.7 (20.3- 23.4) 21.9 (19.4- 23.5) 0.40 21.7(19.8- 23.3) 22.0 (19.4- 23.4) 0.68
Cuff pressure
checks (median
(IQR))
2.9 (1.9- 3.5) 1.7 (0.7- 3.0) 0.041 2.6 (2.1- 3.7) 1.7 (0.7- 3.0) 0.013
Ventilator
circuits checks
(mean (SD))
17.3 (5.2) 17.0 (6.7) 0.88 17.2 (3.8) 17.0 (6.8) 0.94
Summary of the significant explanatory variables found in the study
Table 5.10 illustrates variables in the retrospective study which were found to be significant in
relation to VAP and VAE occurrences. Note that, although underlying disease was found to be
significant at p=0.045, this variable cannot be modelled for VAE as no patients with underlying
respiratory disease were identified as having VAE.
Table 5.10: Explanatory variables with a significant association with VAP or VAE by univariate
analysis
Response variables/ Explanatory variables with a significant p-value (measurements)
VAP VAE
• Duration on mechanical ventilation
(continuous)
• Length of PICU stays (continuous)
• Length of hospital stays (continuous)
• Cuff pressure checks (continuous)
• Underlying disease (cardiovascular, respiratory,
others)
• Duration on mechanical ventilation (continuous)
• Length of PICU stays (continuous)
• Length of hospital stays (continuous)
• Reintubation (yes/no)
• GI prophylaxis (yes/no)
• Cuff pressure checks (continuous)
109
Discussion
Possible risk factors and compliance of VAP preventative strategies
In this study, the reintubation procedure was performed 26.0% (68/262 mechanical ventilation
episode). This was due to planned but failed extubation, accidental extubation or when ETT change
was deemed clinically necessary. This proportion of reintubation procedure occurred was higher
compared to a prospective study by Gautam et al. (2012) who reported 7.1% of 269 patients have
been reintubated. Elward et al., (2002) in their study identifed only 13.7% of 911 patients were
undergone reintubation procedure.
Reintubation increases the likelihood for microaspiration of oropharageal secretion that could lead to
VAP. The majority of patients in this study were nasally intubated, were less sedated and were
administered muscle relaxants during the course of mechanical ventilation. These factors are also
noted in previous studies as factors that could explain the relationships with VAP development
(Coffin et al., 2008; Hellyer et al., 2016; Zolfaghari & Wyncoll, 2011). While almost all patients had
a nasogastic tube while receieving mechanical ventilation, nearly 50% of the patients received GI
prophylaxis, reflecting common practise in PICU patients. This was also reported by previous studies
(Albert et al., 2016; Heyland et al., 2004; Prakash et al., 2016).
The overall compliance to VAP preventive strategies performed by nurses for ventilated patients in
the 2015 cohort was 89%, showing that compliance was at acceptable ranges according to Tabaeian
et al. (2017), but lower than the benchmark (above 95%) set by the IHI (Resar et al., 2014). This
compliance rate is similar to other compliance rates reported in paediatric studies by Brierley et al.
(2012) and Bigham et al. (2009) but lower than rates from De Cristofano et al. (2016) (above 95%).
The overall compliance rate is considered to be within an acceptable range given the challenges in
implementation and monitoring due to the transitioning of two former PICUs into one at QCH. This
data, however, forms the baseline data for the unit (PICU).
Hand hygiene and frequency of endotracheal suctioning met 100% compliance. One possible reason
is the raised awareness of the importance of hand hygiene in healthcare organisations (Bouadma,
Mourvillier, Deiler, Le Corre, et al., 2010). Another possible reason includes the mandatory
monitoring of the effectiveness of the National Hand Hygiene Initiative which is measured monthly
through auditing conducted by hospital staff (Hand Hygiene Australia, 2017b).
The suggested frequency of ETT suctioning is four times per day (see Chapter 4, Table 4.1). This
frequency is based on patient stability and is highly dependent on clinical assessment of need. Given
110
that many children are intubated for an underlying respiratory disease with subsequent increased
secretions it is not surprising that 100% compliance is achieved (Davies et al., 2011; Morrow &
Argent, 2008). Other VAP preventative strategy compliance, such as oral hygiene, cuff pressure
checks and HOB elevation were reported above 80%. This may reflect the standard of nursing care
and baseline knowledge of unit staff.
The lowest compliance reported (70.8%) was for ventilator circuit checks. However, there may be a
gap between performing a task and documentation. All data in this study is derived from documented
procedures as provided by the hospital.
Discussion of the association between study variables and VAP/VAE development at
univariate analysis level
The duration of mechanical ventilation, length of PICU and hospital stays were associated with the
development of VAP and VAE. These findings corroborated other studies (Awasthi et al., 2013;
Becerra et al., 2010; Cocoros et al., 2016; Elward et al., 2002; Hatachi et al., 2015; Iosifidis et al.,
2016; Phongjitsiri et al., 2015; Roeleveld et al., 2011; Tang et al., 2009). This suggests that patients
on mechanical ventilation are exposed to ventilator-associated complications (VAP and VAE), and
that the impact of this increases hospitalisation duration.
However, the development of VAP and VAE were not directly associated with mortality. This is
consistent with some contemporary VAP studies (Almuneef, Memish, Balkhy, Alalem, & Abutaleb,
2004; Elward et al., 2002; Gautam et al., 2012; Iosifidis et al., 2015; Srinivasan et al., 2009). In
contrast to the present study, VAE was found to be associated with hospital mortality, as shown by
Cocoros et al. (2016); Iosifidis et al. (2016); Phongjitsiri et al. (2015). This possibly suggests that
there is no significant association between VAP/VAE incidence and hospital mortality rate, but this
is affected by relatively low VAP and VAE rates. In this study only three patients who met both VAP
and VAE surveillance tools died, but this finding could not claim that the mortality was due to
VAP/VAE.
The present study found that only ETT cuff pressure checks were associated with the development of
VAP in univariate analysis. This may be explained by possible aspiration of pathogenic secretions to
the lower respiratory tract (Dave et al., 2011) due to inadequate cuff pressure. In this study, some
patients clinically required a change from an uncuffed to a cuffed ETT. More than 90% of patients in
the 2015 cohort were mechanically ventilated with a light sedation state which still permits patient
111
movement. This movement may also contribute to cuff leakage and consequently there is the potential
for oropharyngeal secretions’ aspiration to occur (Mietto et al., 2013).
Reintubation, high acuity (PIMS3) score, presence of nasogastric tube and nasogastric feeding,
female gender and post-surgical state have been previously shown to be associated with VAP
development in children (Awasthi et al., 2013; Casado et al., 2011; Elward et al., 2002; Gupta et al.,
2015; Kusahara et al., 2014; Roeleveld et al., 2011). These risk factors were not identified during the
present study. Reintubation was not a risk factor for VAP in the present study, but this may be due to
75% of patients not requiring reintubation. The majority of children were nasally intubated which
could reduce the chance of unplanned extubations and possible reintubations (Gupta & Rosen, 2016).
These subsequently may reduce the potential for microaspiration of pathogenic secretion in to the
lungs that could lead to VAP (Berry, Davidson, Masters, et al., 2011; Dave et al., 2011).
The present study found that reintubation, cuff pressure checks, and the presence of GI prophylaxis
were associated with the occurrence of VAE at univariate analysis (p<0.05). Previous studies have
identified neuromuscular blockage, blood transfusion, positive fluid balance, tracheostomy,
immunocompromised status, chronic respiratory disease, mean peak inspiratory pressure, acute
kidney injury, and trauma as being associated with VAE in children (Beardsley et al., 2016; Cocoros
et al., 2016; Cocoros, Priebe, Gray, et al., 2017; Guess et al., 2018; Phongjitsiri et al., 2015). This is
the first study that has found an association between VAE development and reintubation, cuff
pressure checking and the presence of GI prophylaxis.
For cuff pressure checks and VAE development, this result may be explained by current practises
which dictate the frequency at which ETT cuffs should be checked. Although the compliance to cuff
pressure checking (twice daily) in this cohort was satisfactory (despite no universal accepted range
for frequency of cuff pressure checks), a huge variation in cuff pressures was reported over the
duration of mechanical ventilation in patients with cuffed ETTs (Memela & Gopalan, 2014), possibly
due to air leakage. Optimum cuff pressure for children is unknown, with both under- and overinflated
cuffs exposing the patient to air leakage which may compromise oxygenation and increase the
potential for microaspiration.
The present study found an association of GI prophylaxis and VAE incidence, which is consistent
with the adult study by Klompas et al. (2016) who also examined the PVAP tier in the VAE
subcategories. Although no relationship was found between GI prophylaxis and VAP in the present
112
study (using the PNU1/VAP surveillance tool), the finding suggests that using the new VAE
surveillance tool in children may potentially identify other complications.
Multivariate analysis
Incidence rate ratios per hour of ventilation (IRR) for each individual risk factor/preventative strategy
are presented in Tables 5.11 and 5.12. These are adjusted incidence rates accounting for age and
gender, which is commonly seen in the analysis of VAP and VAE (Klein Klouwenberg et al., 2014;
Klompas, Kleinman, et al., 2014; Muscedere et al., 2013; Patria et al., 2013; Phongjitsiri et al., 2015).
Only the duration of mechanical ventilation is included in the models. Length of PICU stay and length
of hospitalisation are excluded due to collinearity.
VAP models
Age and gender adjusted Poisson and negative binominal models
All risk factors and preventative strategies with a p-value of 0.15 or less were examined. The results
show some large effects, for example 1.40, however the 95% confidence intervals (CIs) in the
modified Poisson model for reintubation episodes, paralytic agent, gastrointestinal prophylaxis,
routes of intubation and oral hygiene are wide, indicating that these IRRs, although large, are not
robust estimates of the ability of these variables to predict the incidence of VAP per hour of
ventilation. The natural log alpha parameter in the negative binominal regression model also reveals
a large degree of dispersion for hand hygiene, oral hygiene and HOB elevation with the 95% CI
include 1, indicating this model may not be a good fit for this data.
Only cuff pressure checks were found to be significant in the modified Poisson and negative
binominal models (p=0.096 and 0.091) based on the criteria for variable selection for the multi-
variable models (p ≤ 0.15). The negative binomial regression appears to be overdispersed as indicated
by the lack of the 95% CI for log alpha (-14.43–11.40). Thus, this preventative strategy is not an
important predictor of the incidence of VAP per hour of ventilation, at the 5% level of significance.
VAE models
Age and gender adjusted Poisson and negative binominal models
All risk factors and preventative strategies were examined in VAE models, except for nasogastric
tube presence and sedation level as there were empty cells identified during the univariate analysis.
Weight was also excluded because weight is correlated with age although weight was identified as a
potentially important predictor of VAE at univariate analysis.
113
The age and gender adjusted in the Poisson regression model appears to have a large effect size (2.15;
1.26; 4.95; 0.74; and 1.84) with a wide 95% CI for reintubation episodes, paralytic agent,
gastrointestinal prophylaxis, routes of intubation and oral hygiene (0.76–6.67; 0.37– 4.31; 1.09–
22.49; 0.20–2.74 and 1.28–2.64). These results indicate that the incidence rate ratios, although large,
are not robust estimates of the ability of these variables to predict the incidence of VAP per hour of
ventilation. The natural log alpha parameter which is a parameter in the negative binominal regression
models was significant for reintubation episodes and head of bed (HOB) elevation (95% CI includes
1), indicating that for these two variables there is overdispersion present and the Poisson model may
not be a good fit for this data.
114
Table 5.11: Age and gender adjusted Poisson and Negative Binominal regression models of risk factors and preventative strategies for incidence
of VAP/hour of ventilation
Modified Poisson Negative Binomial
Explanatory variables N Age and gender
adjusted IRR
(95% CI)
p-value Age and gender
adjusted IRR
(95% CI)
p-value lnalpha
Reintubation episodes (yes) 262 1.16 (0.43 - 3.13) 0.77 1.25 (0.33 - 4.77) 0.75 -0.41 (-7.24 - 6.42)
Paralytic agent (yes) 262 1.40 (0.45 - 4.31) 0.56 1.40 (0.45 - 4.31) 0.56 -9.57 (-34.51 - 15.37)
Gastrointestinal prophylaxis (yes) 262 1.25 (0.44 - 3.54) 0.68 1.29 (0.40 - 4.16) 0.67 -0.82 (-9.19 - 7.54)
Routes of intubation (oral) 262 1.34 (0.41 - 4.35) 0.63 1.34 (0.40 - 4.47) 0.64 -3.46 (-113.50 -106.57)
Sedation level (deep) 262 0.99 (0.12 - 8.45) 0.99 0.99 (0.11 - 8.65) 0.99 -3.00 (-65.22 - 59.23)
Hand hygiene 244 1.02 (0.97 - 1.08) 0.42 1.02 (0.97 - 1.08) 0.43 -12.28 (-15.57 - -9.00)
Oral hygiene 262 0.77 (0.49 - 1.20) 0.25 0.77 (0.49 - 1.20) 0.25 -11.51 (-16.04 - -6.98)
Endotracheal suctioning 262 1.01 (0.85 - 1.22) 0.88 1.01 (0.85 - 1.21) 0.88 -2.54 (-40.57 - 35.48)
Head of bed elevation 262 1.12 (0.95 - 1.31) 0.17 1.12 (0.95 - 1.31) 0.17 -11.80 (-16.96 - -6.65)
Cuff pressure checks 251 1.15 (0.98 - 1.37) 0.096 1.16 (0.98 - 1.37) 0.091 -1.51 (-14.43 - 11.40)
Ventilator circuits checks 262 1.03 (0.95 - 1.12) 0.51 1.03 (0.95 - 1.12) 0.51 -10.40 (-36.36 - 15.55)
IRR=incidence rate ratio/hour of ventilation. CI=confidence interval.
115
Table 5.12: Age and gender adjusted Poisson and Negative Binominal regression models of risk factors and preventative strategies for incidence
of VAE/hour of ventilation
Modified Poisson Negative Binomial
Explanatory variables
N Age and sex adjusted
IRR
(95% CI)
p-value Age and sex adjusted
IRR
(95% CI)
p-value lnalpha
Reintubation episodes (yes) 262 2.15 (0.76 - 6.67) 0.15 3.47 (1.06 - 11.34) 0.039 1.44 (0.22 - 2.66)
Paralytic agent (yes) 262 1.26 (0.37 - 4.31) 0.71 1.42 (0.34 - 6.02) 0.63 1.17 (-0.86 - 3.19)
Gastrointestinal prophylaxis (yes) 262 4.95 (1.09 - 22.49) 0.038 7.00 (1.58 - 31.14) 0.011 1.24 (-0.15 - 2.63)
Routes of intubation (oral) 262 0.74 (0.20 - 2.74) 0.65 0.62 (0.14 - 2.71) 0.52 1.15 (-0.66 - 2.97)
Hand hygiene 244 1.00 (0.96 - 1.05) 0.92 1.00 (0.95 - 1.05) 0.99 1.01 (-0.81 - 2.82)
Oral hygiene 262 1.84 (1.28 - 2.64) 0.001 1.94 (1.38 - 2.73) <0.001 0.82 (-1.17 - 2.82)
Endotracheal suctioning 262 0.91 (0.78 - 1.06) 0.23 0.91 (0.77 - 1.08) 0.30 0.85 (-2.11 - 3.80)
Head of bed elevation 262 1.02 (0.90 - 1.16) 0.72 1.02 (0.90 - 1.17) 0.71 1.04 (0.97 - 3.05)
Cuff pressure checks 251 1.19 (1.03 - 1.36) 0.016 1.23 (1.02 - 1.47) 0.028 1.04 (-0.99 - 3.07)
Ventilator circuits checks 262 1.03 (0.97 - 1.09) 0.40 1.02 (0.96 - 1.09) 0.44 1.01 (-1.13 - 3.15)
IRR=incidence rate ratio/hour of ventilation. CI=confidence interval
116
The final modified Poisson and Negative Binominal models for VAE
Gastrointestinal prophylaxis and oral hygiene were found to be significant predictors of VAE, as
shown in Tables 5.13 and 5.14. The mixed effects Poisson model was not a significant improvement
on the modified Poisson model (likelihood ratio test p=0.06, AIC 134.5 (modified) versus 132.4
(mixed), BIC 152.4 (modified) versus 153.8 (mixed), log-likelihood -62.27 versus -60.18). Both the
modified and mixed effects Poisson models were overdispersed. The negative binomial natural log
alpha value indicates that there was overdispersion present, although this overdispersal disappears
with the mixed effects negative binomial model. The mixed effects negative binomial model was not
an improvement on the negative binomial model (AIC 132.9 versus 132.9 (mixed), BIC 154.3 versus
154.3, log-likelihood -60.44 versus -60.44). The mixed effects Poisson model reported similar
estimates and had a similar fit to the negative binomial models; however, the standard error for
gastrointestinal prophylaxis was very large, which may indicate that the mixed effects Poisson and
negative binomial models are over specified, that is they are too complex for the number of events.
The AIC results favoured the negative binomial model while the BIC favoured the modified Poisson
regression which is conflicting and provides supportive evidence of an over specified model. With
the exception of gastrointestinal prophylaxis, the parameter estimates for all the models are relatively
similar with similar standard errors, p-values and 95% CIs. The model that best fits our data is
therefore the modified Poisson regression model.
Gastrointestinal prophylaxis and oral hygiene were found to be significant predictors of VAE, using
the modified Poisson regression. For every extra oral hygiene given, the incidence of VAE is 2.15
(95% CI 1.27–3.65) times higher per hour of ventilation. For those on GI prophylaxis, the incidence
of VAE is 6.10 (95% CI 1.60–23.21) times higher per hour of ventilation compared to those not
receiving GI prophylaxis. This estimate should be used with caution due to the wide confidence
intervals. This model is not robust due to the low event rate but identifies a risk factor and preventative
strategy which may be influential in the diagnosis of VAE, although this should be re-examined with
a larger study.
117
Table 5.13: Final Poisson models of risk factors and preventative strategies for incidence of VAE/ hour of ventilation
Explanatory variables Modified Poisson Mixed effects Poisson
Beta (SE) IRR (95% CI) p-value Beta (SE) IRR (95% CI) p-value
Gender (female) -0.51 (0.62) 0.60 (0.18 - 2.00) 0.41 -0.57 (0.69) 0.57 (0.15 - 2.19) 0.41
Age (months) -0.00 (0.01) 1.00 (0.99 -1.01) 0.65 -0.00 (0.01) 1.00 (0.98 - 1.01) 0.61
Gastrointestinal prophylaxis (yes) 1.81 (0.68) 6.10 (1.61 - 23.15) 0.008 2.82 (1.23) 16.78 (1.49 - 188.59) 0.022
Oral hygiene 0.77 (0.27) 2.15 (1.27 - 3.65) 0.004 0.87 (0.32) 2.38 (1.27 - 4.43) 0.007
Constant -12.84 (1.46) 0.00 (0.00 - 0.00) <0.001 -15.02 (2.68) 0.00 (0.00 - 0.00) <0.001
Variance attributed to patient n/a 0.00 (0.00) 0
Variance attributed to mechanical ventilation
within an admission n/a 2.52 (2.15) 2.52 (0.47 - 13.43)
IRR=incidence rate ratio/hour of ventilation. CI=confidence interval. n/a=not applicable
118
Table 5.14: Final Negative Binominal regression models of risk factors and preventative strategies for incidence of VAE/hour of ventilation
Explanatory variables Negative binomial Mixed effects Negative binomial
Beta (SE) IRR (95% CI) p-value Beta (SE) IRR (95% CI) p-value
Gender (female) -0.72 (0.63) 0.49 (0.14 - 1.68) 0.25 -0.72 (0.68) 0.49 (0.13 - 1.86)) 0.29
Age (months) -0.00 (0.01) 1.00 (0.98 -1.01) 0.75 -0.00 (0.01) 1.00 (098 - 1.01)) 0.72
Gastrointestinal prophylaxis (yes) 2.70 (0.92) 14.84 (2.47 - 89.31) 0.003 2.70 (1.11) 14.85 (1.67 - 130.68) 0.015
Oral hygiene 1.02 (0.24) 2.76 (1.72 - 4.44) <0.001 1.02 (0.36) 2.76 (1.37 - 5.56) 0.004
Constant -14.45 (1.88) 0.00 (0.00 - 0.00) <0.001 -14.46 (2.49) 0.00 (0.00 - 0.00) <0.001
lnalpha 1.30 (0.53) 1.30 (0.27 - 2.34) 1.30 (0.73) 1.30 (-0.13 - 2.74)
Variance attributed to patient n/a 0.00 (0.00)
Variance attributed to mechanical ventilation
within an admission n/a 0.00 (0.00)
IRR=incidence rate ratio per hour of ventilation. CI=confidence interval. n/a=not applicable
119
Discussion
Discussion of potential risk factors and VAP preventative strategies and association
with VAP/VAE development by multivariate analysis
In this study, the multivariate analyses examined the ability of possible risk factors and VAP
preventative strategies to predict VAP/VAE. The multivariate analyses undertaken included all
potential risk factors and VAP preventative strategies. Only duration of mechanical ventilation was
taken into the model for incidence rate ratios excluding the length of PICU, length of hospitalisation,
age and weight due to collinearity. This collinearity affects the stability of the data (Weisberg, 2005).
It is understood that these variables correlate with each other.
The majority of previous studies have used logistic regression analysis to examine the possible risk
factors for VAP/VAE development (Casado et al., 2011; Elward et al., 2002; Patria et al., 2013;
Phongjitsiri et al., 2015; Roeleveld et al., 2011; Srinivasan et al., 2009). Moreover, some studies
counted patients as experimental units (Roeleveld et al., 2011; Srinivasan et al., 2009) instead of
mechanical ventilation episodes (Elward et al., 2002; Gautam et al., 2012), and some of the studies
ignore the multiple mechanical ventilation episodes in a person with multiple PICU admissions
(Balasubramanian & Tullu, 2014; Gautam et al., 2012). To our knowledge, none of the statistical
modelling methods used in published VAP research take time at risk into account, which may not
allow for meaningful conclusions in predicting possible risk factors/VAP preventative strategies at
an incidence rate ratio of VAP/VAE per hour of ventilation. Poisson and Negative Binomial mixed
effects models allow for multiple mechanical ventilation episodes and multiple admissions and report
incidence rate ratios, bringing new statistical methodology to the field.
It was found in the present study that an increased frequency of oral hygiene performance was
associated with an increase of VAE of incidence by 2.15 times per hour of ventilation. This finding
corroborates the findings of the adult study, which suggested that oral hygiene using chlorohexidine
was associated with increased mortality (Klompas et al., 2016). This result does need to be interpreted
with caution as a wide confidence interval was reported and with a very low event rate.
Oxygenation deterioration in mechanically ventilated children is more likely to occur compared to
adults because of the risk of airway leakage due to the conical shape of the airway in infants and
children (Khine et al., 1997). Moreover, infants and children are at a higher risk of rapid deterioration
because they expend proportionally more metabolic effort with the increased work of breathing
(Gupta & Rosen, 2016). Given that oral hygiene performance in children is risky (especially in non-
sedated paediatric patients), there is the possibility of worsening oxygenation while performing oral
120
hygiene, particularly for oral tubes. Even in critically ill adult patients, the ideal frequency of oral
care is difficult to determine and performing oral hygiene on uncooperative patients may pose a risk
for aspiration which may subsequently compromise oxygenation (Ames, 2011). The method in which
oral care was performed in the study by Klompas et al. (2016) and how this increased the risk of VAE
development was unclear, but a finding from their metanalysis suggested that it may be due to the
aspiration of chlorohexidine which could cause lung injury (Klompas, Speck, Howell, Greene, &
Berenholtz, 2014).
Another finding of this study noted that patients who received GI prophylaxis had a risk rate 6.10
times higher/hour for VAE. Nevertheless, a wide confidence interval with a low event rate was
reported and thus the interpretation should be undertaken cautiously. Similarly to oral hygiene and
VAE, GI prophylaxis was not found to be covered in depth in any current paediatric VAE literature
to date. However, an adult study assessing the VAP bundle component versus VAE surveillance tools
was found and used as a reference in this study. Through this data, it was found that GI prophylaxis
was associated with risk for VAE (PVAP-tier) [hazard ratio, 7.69; (95%CI, 1.44 -41.10; p =0.02)]
(Klompas et al., 2016). Interestingly, the GI prophylaxis previously was found to be significant at
both univariate and multivariate analysis, which may further support the need for further research
regarding the mechanism of action in children (Albert et al., 2016) and the usefulness of VAE
surveillance tool in children.
Summary
In this chapter, Phase 1: Retrospective study provided detailed epidemiological data for VAP and
VAE in the 2015 cohort of PICU patients in the QCH. The results of VAE using the VAE surveillance
tool described incidence rates, possible risk factors, and the existing VAP preventative strategies for
critically ill children. The VAE surveillance tool in this cohort of patients had high specificity, which
led to a broader explanation of other possible ventilator-associated complications. The compliance of
VAP preventative strategies was measured at 89.0%. Modelling of VAP and VAE was completed but
the findings are not robust due to low numbers of VAP and VAE in this study. They do, however,
provide methodology on which to build future work in this field. None of the risk factors or VAP
preventative strategies were found to be predictive of VAP at multivariate analysis. Oral hygiene
performance and the presence of GI prophylaxis were identified as potentially important predictors
of VAE, which mirrors the latest adult study assessing VAP preventative strategies using the new
VAE surveillance tool.
121
Chapter 6 presents the results and discussion for Phase 2: VAP preventative strategy compliance
auditing and surveys on ‘Speaking up for hand hygiene’.
122
Results and discussion Phase 2: VAP preventative strategy compliance and
surveys
Introduction
This chapter describes and interprets the findings of Phase 2. The aim of this phase was to assess the
compliance auditing of the VAP preventative strategies in the PICU at Queensland Children’s
Hospital. Nursing staff and parents’ opinions of the ‘Speaking up for hand hygiene’ initiative was
also assessed. This chapter addresses the following research questions:
1. What are the current VAP preventative strategies in Queensland Children Hospital PICU?
2. What are the perceptions of parents and nurses in Queensland Children Hospital PICU of the
‘Speaking up for hand hygiene’ component of the VAP education?
The methodology for this phase is described in Chapter 4, Section 4.3.2.
Results of compliance auditing of VAP preventive strategies
Demographic characteristics of participants in VAP preventative strategy compliance
auditing
Compliance auditing was undertaken to describe the current status of VAP preventative strategy
compliance in PICU. The overall compliance of VAP preventative strategies (VAP bundle) was
83.1%. The audit involved a total of 183 individuals who were responsible for the care of 37 patients;
a total of 204 mechanical ventilation days. Of the 183 individuals, 84 were nurses, 50 medical
practitioners, 12 allied health personnel, and 37 parents. Twenty-eight of the patients were male and
nine were female. The median age of these patients was five months (IQR 0.9–36.0) and the median
number of mechanical ventilation days was four (IQR 2.0–7.5).
Hand hygiene
Both PICU staff and parents were audited for hand hygiene compliance. Of the 475 instances of hand
hygiene observation of nurses, 410 hand hygiene observations were compliant, resulting in
compliance rate of 86.3%. Medical practitioner hand hygiene compliance was measured at 87.4%
(97/111 observations), allied health personnel 86.1% (31/36 observations), and parents 64.7% (44/68
observations).
123
Oral hygiene
Oral hygiene data from a total of 204 mechanical ventilation days was evaluated. The compliance in
performing 12-hourly oral health assessments was at 84.3% (172/204 days). The adherence to
frequency of oral hygiene performance and age-appropriate oral care practise using a recommended
cleaning solution as per unit guidelines was at 52.9% (108/204 days) (Figure 6.2).
Cuff pressure check
The rate of compliance for 12-hourly cuff pressure checks was 70.9%, with practise compliance found
in 107/151 mechanical ventilation days. Cuff pressure check compliance in terms of maintaining the
cuff pressure within limits was 86.1% (130/151 days).
Endotracheal suctioning
Endotracheal suctioning (open method) was performed mainly by nurses and on some occasions by
physiotherapists. A total of 31 episodes of suctioning were observed. Adherence to aseptic non-touch
techniques during suctioning was 74.2% (23/31). Full compliance (100%) was recorded for the
practise of connecting the filtered test lung to the ventilator circuits, after disconnection of ventilator
circuits from the endotracheal and the use of a single suction catheter while performing endotracheal
suctioning (Figure 6.1). The practise of draining the condensate away from the patient and/or
expelling the condensate onto a disposable cloth before re-connection to endotracheal tube showed a
compliance rate of 22.6% (7/31).
Figure 6:1: Endotracheal suctioning (open method) compliance
100.0%
100.0%
74.2%
22.6%
Single-use catheter during endotraheal tube
suctioning (open method)
Connecting the ventilator circuits to filtered test
lung prior re-connection to endotracheal tube
Aseptic non-touch technique adherence
Draining condensate of ventilator circuit away
from patient or expel onto disposible cloths
before re-connection to endotracheal tube
Endotracheal tube suctioning (open method) compliance
124
Head of bed (HOB) elevation
HOB elevation compliance was measured at 86.4%. HOB compliance was reported for a total of 204
mechanical ventilation days.
Ventilator circuits checks
The compliance rate for changing the expiratory bacteria filter every 24 hours recorded during
auditing was 95.2%. The changing of the expiratory filter every 24 hours was observed in 60/63
mechanical ventilation days.
Enteral feeding commencement within 24 hours of admission
Full compliance (100%) was recorded for enteral feeding commencement in 29 patients. Eight
patients were excluded from the analysis because they had either a medical or surgical
contraindication for feeding commencement.
Discussion of compliance of VAP preventative strategies during auditing
The overall compliance rate of VAP preventative strategies (VAP bundle) during the auditing was
83.1%, indicating that compliance was below the expected level (95%) set by the Institute for
Healthcare Improvement (IHI) (Resar et al., 2014). Nevertheless, overall compliance in this phase is
classified as acceptable (75-100%) according to Tabaeian et al. (2017). The inclusion of hand hygiene
within the VAP bundle may explain failure to achieve the IHI benchmark.
The inclusion of hand hygiene in the VAP bundle may lead to results that do not fully represent
compliance because it is not a direct patient intervention (Resar et al., 2014). It appears that when
hand hygiene is included, overall compliance tends to drop below 95%, as cited by Bigham et al.
(2009) and Brierley et al. (2012). De Cristofano et al. (2016) reported compliance above 95% in their
VAP preventative strategies only when excluding hand hygiene compliance assessments. Overall
compliance to the VAP preventative strategies in the present study fits within the acceptable range
since the re-launching of updated VAP education to PICU staff. This is the first time the overall
compliance of VAP preventative strategies and individual VAP preventative strategies was audited
in the unit.
At the level of individual VAP preventative strategies, hand hygiene compliance among PICU staff
(nurses, medical practitioner and allied health staff (physiotherapists)) was reported as being in line
125
with the Australian national benchmark (>80%) (Hand Hygiene Australia, 2017b). It is also consistent
with results of monthly compliance hand hygiene auditing conducted by hospital staff in the PICU
achieving 86% compliance (March–April 2017) (Children’s Health Queensland Hospital and Health
Service, 2017). This result was defined as a high level of hand hygiene compliance by Sickbert-
Bennett et al. (2016).
The findings above agree with those from a study by Bigham et al. (2009) where an improvement of
hand hygiene compliance from 60% to 90% was reported, although they specifically evaluated hand
hygiene practise before and after contact with ventilator circuits. Sickbert-Bennett et al. (2016)
reported that the hand hygiene compliance of healthcare workers was 95% in the PICU following the
implementation of their ‘Clean In, Clean Out’ program which asks that all staff perform hand hygiene
every time they enter and exit a patient room. The present study used similar hygiene compliance
assessments as those undertaken in study by Sickbert-Bennett et al. (2016), where ‘5 Moments of
Hand Hygiene’ were assessed. Another study that found incremental increases in hand hygiene from
65% at baseline to 88% in Neonatal ICU staff after a hand education program was implemented
(Helder et al., 2010).
This high level of hand hygiene compliance reported in the present study may be explained by a few
factors including staff engagement with patient safety and ongoing organisational hand hygiene
auditing. Another factor which may have contributed to these rates was the launching of a video called
‘It is OK to ask me to wash my hands’. Previous studies have demonstrated that the compliance of
hand hygiene among healthcare workers may also be influenced by the actions of colleagues and the
effort made by the organisation/unit to promote a positive culture (Boscart, Fernie, Lee, & Jaglal,
2012; Dixit, Hagtvedt, Reay, Ballermann, & Forgie, 2012; Erasmus et al., 2010).
There is also the possibility of the Hawthorne effect, especially in relation to hand hygiene
compliance assessments (Chen et al., 2013; Dhar et al., 2010). The Hawthorne effect refers to the
tendency of an individual who is being observed to alter their behaviour towards expected practise
(McCambridge, Witton, & Elbourne, 2014; Parsons, 1974). The present study minimised this
possibility by performing unobtrusive hand hygiene auditing at random times (Rosenthal, Alvarez-
Moreno, et al., 2012). The monthly hand hygiene compliance audits run by trained hospital staff were
held separately from the study compliance audits. These results (not presented) were similar to those
obtained by the researcher, demonstrating that measurement bias was minimised (Chen et al., 2013).
126
A unique aspect of this study was the measuring of hand hygiene compliance by parents. Although
compliance reached above 60%, this result could not be compared to any benchmark of hand hygiene
compliance because there is none in a PICU setting (World Health Organization (WHO), 2009a). One
study found hand hygiene compliance amongst parents and family members in a Neonatal ICU to be
on average 71%. This was not considered an acceptable compliance rate by the Prevention Committee
(IPC) of the hospital (Chandonnet et al., 2017). These findings highlight that parents, as a target group
to enhance hand hygiene compliance in the prevention of hospital-associated infections (HAIs), need
a good standard of education for hand hygiene (Anthony et al., 2013; Chandonnet et al., 2017; World
Health Organization (WHO), 2009a).
Two elements of oral hygiene during auditing reported a compliance gap. These are: compliance with
12-hourly oral hygiene assessment, which was 84%; and frequency of age appropriate oral hygiene
practise compliance, which was 50%. Theoretically, the assessment of oral hygiene should be
followed with compliance with age-appropriate oral care practise (Bigham et al., 2009; Brierley et
al., 2012; Johnstone et al., 2010). In QCH, four-hourly oral hygiene is expected in mechanically
ventilated children. However, this oral hygiene compliance was consistent with a study by Bigham et
al. (2009), which observed the practise at a baseline of 60%. Other possible reasons that have an
impact on the quality of oral hygiene undertaken in ICU are:
(1) inadequate time
(2) prioritising of other tasks over oral hygiene
(3) perception that oral hygiene is unpleasant
(3) lack of knowledge of oral care for intubated patients (Allen Furr, Binkley, McCurren, & Carrico,
2004).
The ETT cuff pressure monitoring measured the compliance rate of maintaining cuff pressure within
limits set by the unit and results were found to be in line with the study by Bigham et al. (2009). The
cuff pressure was set to a minimum 10 cmH2O, and maximum 20 cmH2O by the study setting. This
is in line with the parameters used by Weiss et al. (2009) and Rosenthal, Alvarez-Moreno, et al.
(2012). This pressure is believed to be able to provide an adequate seal to avoid aspiration of
pathogenic secretions into the lower respiratory tract (Dave et al., 2011; Bhardwaj, 2013).
Full compliance of frequency of ETT suctioning was found. This result is expected, as the frequency
of ETT suctioning varies across patients and relies on clinical indicators (Morrow & Argent, 2008).
The Hawthorne effect could be one of the factors contributing to this result. Compliance of aseptic
non-touch technique for endotracheal suctioning, and the practise of draining the condensate
127
ventilator circuit away from patients was 74% and 23% respectively. These findings may suggest
potential for contamination to occur which may in turn facilitate VAP occurrence (Coffin et al., 2008;
Kollef, 2004; Tolentino-DelosReyes et al., 2007). This discrepancy is likely to be related to practise
variations, although a sound knowledge of aseptic technique by nurses has been reported (Leong et
al., 2017). Full compliance was achieved for the practise of using single-use catheters during ETT
suctioning and connecting the ventilator circuits to a filtered test lung. Full compliance in these
elements may help minimise the risk of cross-infections (Tolentino-DelosReyes et al., 2007).
The compliance to HOB elevation was 86%. In this study, HOB was set at 15–30 degrees (Queensland
Children's Hospital Paediatric Intensive Care Unit, 2016), which is similar to a previous study, which
used a range of 20–30 degrees (Brierley et al., 2012). A compliance rate of 85% was also reported in
a study by Bigham et al. (2009), where the elevation was set at 30–45 degrees for paediatric patients.
More than 95% compliance was reported for ventilator circuit checks, measured by the change of
expiratory bacteria filter every 24 hours. This was higher than the 85% compliance rate reported in a
previous study (Bigham et al., 2009). The possible reason for the high compliance found in the present
study could be the requirement to change the filter every 24 hours either by the in-charge nurse or
respiratory therapist scheduled before 4.00pm every day. This preventative measure helps to
minimise bacterial contamination (Klompas, Branson, et al., 2014; Resar et al., 2014).
Full compliance for early commencement of enteral feeding within 24 hours of admission (in cases
with no contraindication) was reported during auditing. This compliance was anticipated because it
is routinely assessed after PICU admission and reviewed regularly thereafter during patient rounding
in this study setting.
Parental survey: ‘Speaking up for hand hygiene’
Response rate
VAP education was delivered to 30 parents during the data collection period. Each parent opted for
a hard copy of the survey as opposed to an online version. One parent was excluded due to a language
barrier. There were 11 losses to follow up. Of those, five patients were discharged home without the
parents returning the survey, two parents felt hesitant to continue the survey and decided not to
participate, two children died, and two children had a deteriorating condition while in treatment. A
total of 19 surveys was returned, constituting a 63.3% response rate.
128
Demographic characteristics of parents and their child’s admission history
A total of 19 parents returned the survey, however some of them did not filled three demographic
characteristics. This was (i) education level of parents (2/19), (ii) history of child their being admitted
to PICU (1/19) and (iii) history of their child experiencing/receiving mechanical ventilation (1/19).
Majority of parents who participated in the survey were female 15 (78.9%). Most parents (12; 63.2%),
were 31 years old or above. Of the 17 parents who responded to the survey, there was an equal
proportion who had formal qualifications (nine; 52.9%), and those who did not (eight; 47.1%). Most
parents were not employed in the healthcare field (14; 73.7%). Of 18 parents, only two (11.1%)
reported that their child had a previous history of a PICU admission and 14 (77.8%) had no prior
experience of their child receiving mechanical ventilation. Eleven (57.9%) parents reported that they
had heard about VAP. Most of the parents preferred to perform hand hygiene using either hand gel
or by hand washing (13; 68.3%), while four (21.1%) used hand wash only, and two (10.5%) used
hand gel only.
Perceptions of parents on VAP education
Hand hygiene was perceived by parents as important in the prevention of infection in hospitals,
including VAP in PICU. All parents felt hand hygiene was important for nurses and other PICU staff
providing care to their child. Most parents agreed that nurses (18; 94.7%) and other PICU staff (17;
89.5%) wash their hands enough in PICU. Of 19 parents, 10 (52.6%) agreed that the pamphlet, ‘VAP:
How I Can Help my Child in PICU’ was easy to understand. Of the 19 parents, 10 (52.6%) disagreed
that the information in the pamphlet made them concerned and nine (47.4%) parents perceived that
the information was upsetting.
Parental perceptions on willingness to remind and to be reminded by nurses and other
PICU staff to perform hand hygiene
Most parents reported that they would be willing to remind nurses (13; 68.4%), and other PICU staff
(doctor, physiotherapist, dietitian, occupational therapist) (15; 78.9%) to perform hand hygiene when
necessary. Most parents (18; 94.7%) agreed that they were willing to be reminded of hand hygiene
by nurses. Parents agreed that the ‘Speaking up for hand hygiene’ initiative would increase hand
hygiene practises amongst nurses (15; 78.9%) and other PICU staff (14; 73.7%).
Reasons parents would be reluctant to prompt nursing staff regarding hand hygiene
Parents identified several reasons that would make them reluctant to prompt nursing staff regarding
hand hygiene. Eight (38.1%) responses from parents felt that it was not their place to remind the
129
nurses to perform hand hygiene, with one parent (4.8%) in a free written response highlighting their
concern that if they remind the nurses, the attitude of nurses might change. Five (23.8%) responses
from parents noted that they worried that if they reminded nurses, it would affect the care of their
child and they did not want to interrupt nurses by reminding them to perform hand hygiene. Two
parents reported that they would be too embarrassed to remind the nurses (9.5%).
Reasons parents would be reluctant to remind other PICU staff regarding hand
hygiene
Parents provided several responses as to why they would be reluctant to remind other PICU staff of
hand hygiene. Ten (43.5%) responses indicated that they felt it was not their place to remind other
PICU staff to perform hand hygiene with one (4.3%) worried about changing the attitude of PICU
staff if parents elected to remind them of hand hygiene. Four (17.4%) responses respectively
represented parents feeling worried that if they reminded other PICU staff, it would affect the care of
their child, or that they did not want to interrupt other PICU staff by reminding them to perform hand
hygiene and would be too embarrassed to remind the other PICU staff.
Suggestions or comments to improve hand hygiene practise in the unit
Eight of the 19 parents made suggestions and comments to improve hand hygiene within the unit.
The majority of these suggestions focussed on clear communication to convey good hand hygiene
practise. Suggested communication included verbal reminders directly to the nurses and other PICU
staff before coming into contact with their child. They suggested that nurses and other PICU staff
should strictly enforce hand hygiene with parents and visitors. Parents also suggested that more effort
should be made by the unit to make parents feel comfortable to remind the staff to perform hand
hygiene. Visual reminders were also suggested, with examples such as signs with images of germs
and additional floor signs visible upon entry into a room in case wall signs are not seen.
Discussion of parental perceptions on ‘Speaking up for hand hygiene’
The survey response rate from parents was 63.3%. The parents’ response rate is quite similar to that
reported in other surveys involving family members, with response rates of 63.8% and 64.7% by Pan
et al. (2013) and Wu et al. (2013) respectively. However, these two studies were conducted across
multiple sites compared to a single hospital unit involved in the present study. Due to the time
constraints of having only three months for the survey (Phase 2), the invitation for the survey was
terminated although the target number of participants was not achieved, based on the initial sample
size calculation (78 parents). However, measures have been done to increase the response rate. For
130
example, five parents were followed up when they were not returning the survey after discharged
home. In these cases, online version was offered, yet no response received. Two parents that their
children were discharged to ward also were followed up daily and arrangement with ward staff nurse
also done to collect the survey when they completed the survey. Despite that, they felt hesitant to
continue the survey.
Most participants were mothers, congruent with other studies, such as that by Buser, Fisher, Shea,
and Coffin (2013), who reported 80% of their participants as mothers. Parents used the combination
hand hygiene (hand gel and hand wash) in the PICU, which reflects the results of Ciofi degli Atti et
al. (2011). A small fraction of parents preferred hand wash over the use of an alcohol-based rub, a
finding consistent with another study involving general healthcare workers. This preference may be
influenced by a number of factors, including knowledge regarding the benefits of this method or other
factors preventing use, such as those with allergies (Karaaslan et al., 2014). This finding signals the
importance of making educational resources and hand hygiene resources (both alcohol-based rub and
soaps) available in intensive care units.
Parents perceived VAP education delivered through the pamphlet ‘VAP: How I Can Help my Child
in PICU’ as easy to understand. Education to promote hand hygiene among parents and patients
requires information to be presented in a way that enhances understanding. This includes a range of
resources and careful consideration with language and formatting that is suitable for the lay audience.
A similar approach is noted by Davis, Parand, Pinto, and Buetow (2015) and Chandonnet et al.
(2017); they used leaflets, information sheets, posters and videos to convey information. In this study,
it was found that the information in the pamphlet provided to parents was easy to understand given
the language used was simple and it was created with input from PICU staff (Australian Bureau of
Statistics, 2009). The pamphlet had been extensively revised by different panels during five validation
rounds before being printed and distributed to ensure the language level and content were appropriate
(see Chapter 4, Table 4.4).
Parents reported that the information about VAP in the pamphlet was important, but it did make them
feel concerned about hand hygiene. Factors which could explain this observation may include that
the parents are in a state of fear and ongoing stress, and hence their concern gravitates more towards
their child’s wellbeing/stability rather than the information related to hand hygiene (Bellissimo-
Rodrigues et al., 2016; Ciofi degli Atti et al., 2011; Uhl, Fisher, Docherty, & Brandon, 2013).
131
Most parents agreed that hand hygiene is important in VAP prevention in PICU. This is in line with
the previous studies where Buser et al. (2013) found that the majority (78%) of parents perceived
hand hygiene as the most important practise needed to prevent hospital infection. Parents agreed that
hand hygiene of nurses and other PICU staff is important while providing care to their child and the
majority agreed that nurses and other PICU staff wash their hands sufficiently while caring for their
child in PICU. This result accords with the earlier study by Pan et al. (2013) which reported that
77.1% of healthcare workers wash their hands adequately according to hand hygiene standards.
Almost all parents in the present study (94.5%) were willing to be reminded by the nurses and other
PICU staff to perform hand hygiene when necessary, but only 68.4% (14/19) of parents were willing
to remind the nurses (14/19) and other PICU staff 78.9% (15/19) to perform hand hygiene. This
difference was also noted in another study by Wu et al. (2013) where 50.8% (425/836) and 48.9%
(410/839) of families were willing to remind nurses and doctors to perform hand hygiene respectively.
The majority of families agreed (96.5%) that they should help remind healthcare workers to perform
hand hygiene, but only 67.2% of them were actually willing to remind the healthcare workers (Pan et
al., 2013). The level of parents’ willingness to remind nurses to wash their hands could be influenced
by social barriers caused by the healthcare workers’ professional status (McGuckin, Storr, Longtin,
Allegranzi, & Pittet, 2011). This is also mirrored in the study by Kim et al. (2015), which found that
70% of families believed that it is not their role to remind healthcare workers to perform hand hygiene.
Parents (78.9% and 73.7%) agreed that ‘Speaking up for hand hygiene’ initiative would increase hand
hygiene practise among nurses and other PICU staff respectively. Despite this, many of the parents
who reported this also said that they would be reluctant to remind nurses and other PICU staff to
perform hand hygiene because they feel it is not their position to question nurses and other PICU
staff. This finding was similar to several studies where parents/families felt uncomfortable about
reminding doctors and nurses to perform hand hygiene, due to the perceived authority of the medical
staff (Ciofi degli Atti et al., 2011; Dixit et al., 2012; Kim et al., 2015; Longtin, Sax, Allegranzi,
Hugonnet, & Pittet, 2009). It seems that there is an issue of power imbalance, ineffective
communication, professional expectations, and relationship problems with healthcare workers that
prevents parents voicing their concerns regarding their child to healthcare workers (Bellissimo-
Rodrigues et al., 2016; Corlett & Twycross, 2006).
Parents’ suggestions to improve hand hygiene practise in PICU overwhelmingly focused on the need
for clearer communication, with suggestions that the unit should increase efforts to help them feel
more able to remind staff to perform hand hygiene. This was consistent with the previous systematic
132
review by Bellissimo-Rodrigues et al. (2016) and Buser et al. (2013), and is also in line with patient-
centred, family-centred care (Kuo et al., 2012; Muething et al., 2012; Rosen, Stenger, Bochkoris,
Hannon, & Kwoh, 2009). Parents/family members are concerned with how healthcare workers might
treat them or their loved one if they were to voice a safety issue with healthcare workers (Coyne et
al., 2013; Smith et al., 2015). Parents agreed with suggestions to include more visual reminders such
as larger graphics to indicate the risks associated with poor hand hygiene in the PICU. The use of a
visual reminder is one of the strategies to promote hand hygiene among patients to healthcare workers
that may also be applicable to parents (Fitzpatrick et al., 2011; Pittet et al., 2011).
Nursing staff survey
Response rate
The survey was available online and self-administered to 150 nurses in PICU during the data
collection period. Twenty-four nurses responded via the online version and 10 nurses responded via
hard copy, resulting in 34 nurse participants completing the survey. The total response rate was
22.6%.
Demographic characteristics of nursing staff participants
All survey participants were female nurses with the median PICU experience of 5.3 years. The
participants routinely used a combination of both hand gel and hand washing in PICU (31; 91.2%).
Nurses’ perceptions of their willingness to remind and be reminded by parents and
other PICU staff to perform hand hygiene
Most nurses agreed they would remind parents and other PICU staff to perform hand hygiene if
necessary and they were willing to be reminded by parents if they themselves did not perform hand
hygiene (30; 88.2%). Only two (5.9%) nurses disagreed concerning reminders if they came from other
PICU staff. The majority of nurses agreed that ‘Speaking up for hand hygiene’ would increase hand
hygiene compliance amongst parents (31; 91.2%) and other PICU staff (29; 85.3%).
Reasons nurses would be reluctant to prompt parents regarding hand hygiene
Thirty responses were recorded for reasons as to why nurses would be reluctant to remind parents to
perform hand hygiene in PICU. More than half of the respondents (20; 66.7%) provided a response
in the ‘other reasons’ free response option, rather than agreeing with one of three fixed responses,
and identified being concerned about parents’ attributes and behaviour as a contributing factor. This
included potential instances when parents are distressed (e.g. emotionally disturbed or aggressive)
133
and in highly sensitive moments, such as after receiving distressing information. Nurses were also
reluctant to remind parents if the parents were known to have been confronting, defensive or
unwilling to engage in the past, and the nurses feared causing offence. In addition, 5 (16.6%)
respondents noted that they simply did not want to interrupt the parents with their child 3 (10.0%)
nurses felt that it would be too embarrassing to remind parents to perform hand hygiene, and 4
(13.3%) felt that it was not their place to question parents.
Reasons nurses would be reluctant to prompt other PICU staff regarding hand hygiene
Nurses provided 28 responses, 13 (46.4%) of which provided the ‘other reason’ free response. These
showed that nurses were reluctant to remind PICU staff about hand hygiene for reasons such as
finding staff unapproachable, irritated or aggressive, and therefore they perceived a lack of
willingness from them to accept reminders. One response highlighted the issue of medical hierarchies
and seniority in the unit; junior staff felt unable to remind seniors, and some felt that senior staff were
not leading by example. There were also responses that suggested nurses did not remind other PICU
staff because they had missed the opportunity at the time and forgot to do so afterwards. Furthermore,
9 (32.1%) responses found that nurses would be too embarrassed to remind other PICU staff to
perform hand hygiene. Nurses also felt that they did not want to interrupt other staff and felt that it
was not their place to question other staff (3; 10.7%).
Suggestions or comments to improve hand hygiene practise in the unit
Fourteen suggestions or comments were received from 34 nurses to improve hand hygiene practise
in PICU. Overall, there was a perceived need for PICU staff to maintain active involvement in hand
hygiene promotion. More feedback from auditing was also welcomed. Nurses highlighted the
importance of being proactive and vigilant with hand hygiene education not only for patient safety
but also staff and visitors’ protection. Other suggestions were to reintroduce mini hand gel bottles
that could clip to nurses’ uniforms for easy access, and to offer rewards for consistent good practise
at a unit level. Continuity of existing hand hygiene promotion in PICU is supported, with suggestions
to include more reminders, such as large graphics to educate those in the unit about the spread of
disease caused by a lack of hand hygiene.
Discussion of nurses’ perceptions on ‘Speaking up for hand hygiene’
The nurses’ response rate was lower compared to the previous study by Kim et al. (2015), where the
response rate was 84%. In the nurses’ survey, more than 90% agreed that ‘Speaking up for hand
hygiene’ would increase hand hygiene among parents and more than 80% among other PICU staff.
134
Eighty-eight percent of nurses surveyed reported a positive perception of the parents’ role in their
child’s health, and more than 90% were willing to be reminded to perform hand hygiene when
necessary. This is contradicted by a study by Kim et al. (2015) where only 31% of nurses reported
that they were willing to be reminded by parents to perform hand hygiene, and 26% of physicians
supported this view. A previous study reported that healthcare workers perceived that a reminder
from parents was less important to their hand hygiene adherence (Ciofi degli Atti et al., 2011).
Possible factors that would contribute to better perception of nurses towards reminders of hand
hygiene from parents and other PICU staff include increased concern of the unit itself regarding hand
hygiene and the willingness to increase hand hygiene compliance (Boscart et al., 2012). The Patient
Safety and Quality Unit in the study setting published a video promoting ‘Speaking up for hand
hygiene’ and added colourful visual hand hygiene reminders during our data collection period. The
video includes PICU staff holding a poster with the message, ‘It is OK to ask me to wash my hands’.
The main reason for nurses’ reluctance to remind parents to perform hand hygiene was due to the
concern about the parents’ attitudes and behaviour (such as parents being in a state of emotional
distress). It may not be appropriate to ask parents to perform hand hygiene in these circumstances.
This is because nurses are more focused on the immediate consequences of families’ safety, especially
parental emotional status, than on reminding families to perform hand hygiene (Boscart et al., 2012).
The findings of these surveys reveal that nurses were reluctant to remind other PICU staff to perform
hand hygiene if they felt that their working colleagues were unapproachable, moody, or not willing
to accept reminders. This was also related to their perception of the attitudes and behaviours of other
PICU staff (Okuyama et al., 2014). Some felt uncomfortable reminding their colleagues because of
the staff hierarchy; being a junior often prevented them from feeling comfortable advising senior
colleagues (Samuel et al., 2012). One driving factor in hand hygiene concerns maintaining staff
wellbeing — for example, if an infection risk exists (e.g. tuberculosis) (Bellissimo-Rodrigues et al.,
2014). Another factor is motivation, which includes the acuity of patient care, social influences, self-
protection, and the use of cues (Okuyama et al., 2014; Smiddy et al., 2015).
From nurses’ perspectives, suggestions to increase hand hygiene practise were related to the active
involvement of PICU and organisational efforts. These results corroborate the findings of a previous
study that utilised a novel multi-modal strategy of education, performance feedback and the use of an
easy-to-use pocket hand rub dispenser which resulted in improved compliance among nurses,
respiratory therapists and medical personnel (Koff et al., 2011). The use of a similar device attached
135
to the scrubs or gown improved hand hygiene among anaesthetists in operating theatres (Koff et al.,
2009). Efforts initiated by the unit were also found to be an innovative approach to increase adherence
to hand hygiene among healthcare workers; the introduction of badges worn by individuals which
prompted staff to wash their hands resulted in a marked increase in hand hygiene (Boscart et al., 2008;
Levchenko, Hufton, Boscart, & Fernie, 2010). Interestingly, offering rewards to those who comply
with hand hygiene has been found to work exceedingly well. Talbot et al. (2013) enacted this
approach using a financial incentive, and their assessment of healthcare workers’ hand hygiene
compliance improved to more than 95%.
Summary
This chapter has presented the results of Phase 2: VAP compliance auditing and parental and nursing
staff surveys. The findings demonstrated acceptable to high compliance with VAP preventative
strategies was being achieved through VAP education with feedback by both PICU staff and parents.
Both parents and nurses were motivated by the ‘Speaking up for hand hygiene’ initiative to improve
hand hygiene practise in the unit, but a gap persists in the willingness to remind each other to perform
hand hygiene. The findings from the parental and nurses’ surveys have challenged existing evidence,
showing parents and nurses having more positive perceptions about being reminded by each other to
perform hand hygiene than has been previously shown. However, caution should be exercised in
drawing firm conclusions based on these findings due to the small sample sizes. Associations between
these two groups were not examined. Furthermore, the perception of other PICU staff towards parents
and nurses on ‘Speaking up for hand hygiene’ was not examined and remains an area with potential
to improve the study.
Chapter 7 presents the results and discussion for Phase 3: Prospective study.
136
Results and discussion Phase 3: Prospective Study
Introduction
This chapter presents the results and discussion of the prospective study and includes a comparison
with the retrospective study results presented in Chapter 5. The aim of the six-month prospective
study (from the 12th of June until the 12th of December 2017) was to evaluate the incidence of
VAP/VAE in the PICU of Queensland Children’s Hospital (QCH) following the implementation of
the VAP education package and compliance auditing. This chapter addresses the following three
research questions:
1. What was the incidence of VAP and VAE in QCH PICU as defined by the PNU1/VAP and
the VAE surveillance tool?
2. What was the compliance of VAP preventative strategies and potential risk factors for
VAP/VAE in QCH PICU?
3. Was there any improvement in VAP/VAE incidence and VAP preventative strategies
compliance after implementation of the VAP education and compliance preventative
strategies auditing with feedbacks?
The methodology used to produce this chapter’s results is described in Chapter 4, Section 4.3.3.
Results
During the six months of data collection for this study there were 120 episodes of mechanical
ventilation meeting the inclusion criteria (see Chapter 4, Section 4.4.3.1) in 110 patients with 115
admissions. The PNU1/VAP and the VAE surveillance tools were applied to the 120 episodes of
mechanical ventilation. The total mechanical ventilator days as defined by the CDC calculation (end
of ventilation) was 922 days across the 120 episodes recorded. The total number of end points until
patients were no longer at risk of VAP/VAE was 860.96 and 838.43 ventilator days respectively.
Baseline demographic characteristics
Demographic characteristics of patient based on PICU admission
Table 7.1 outlines the demographic characteristics of 110 patients from 115 admissions. More than
60% of the patients were male and aged (<one year) with the median weight of six kilograms. The
PICU admissions in this prospective study mostly came from direct admission (44.3%). The majority
of patients (64%) admitted to PICU were categorised under medical discipline for PICU diagnosis
137
category. Twenty-one percent were associated with cardiovascular and respiratory respectively. The
median of paediatric index mortality 3 (PIMS 3) score was at 4.92. These indicate that the overall
probability of death in this cohort was 4.92%.
The demographic characteristics of prospective and retrospective studies were homogenous except
for underlying disease for PICU admission (p=0.035) and the PIMS3 score (p<0.0001). These suggest
that underlying disease responsible for PICU admission in the prospective study were different in
proportion and the prospective cohort had a higher risk of death compared to those in the retrospective
study. These two sets of data were not comparable.
Table 7.1: Demographic characteristics comparison of patients in prospective study versus
retrospective study on PICU admission (n=115; n= 253) respectively
Variables Prospective
n (%)
Retrospective
n (%)
p-value
Gender Male 71 (61.7) 145 (57.3) 0.42
Female 44 (38.3)
108 (42.7)
Age <1 year 70 (60.9) 158 (62.5) 0.44
1- 12 year 34 (29.6) 80 (31.6)
13 and above 11 (9.6) 15 (5.9)
Weight (kg, median
(IQR))
5.9 (3.5- 16.0)
5.8 (3.6- 15.0)
0.71
PICU source of
admission
0.22
OT/ Recovery 40 (34.8) 70 (27.7)
Emergency Department 9 (7.8) 37 (14.6)
Ward (other inpatient
area)
15 (13.1) 37 (14.6)
Direct admission 51 (44.3) 109 (43.1)
PICU Diagnosis
Category
0.51
Medical 71 (61.7) 163 (64.4)
Surgical 38 (33.0) 71 (28.1)
Trauma 6 (5.2) 19 (7.5)
Underlying disease
Trauma/ Injury 6 (5.2) 29 (11.5) 0.035
Cardiovascular 30 (26.1) 94 (37.2)
Neurological 14 (12.2) 21 (8.3)
Respiratory 30 (26.1) 50 (19.8)
Gastrointestinal 8 (7.0) 10 (4.0)
Miscellaneous 27 (23.5) 29 (11.5)
PIMS 3 (median, IQR) 0.0001
PIMS3 risk of death 4.92 (0.49-
3.44)
1.90 (0.70-
5.50)
OT= Operation Theatre; PIMS= Paediatric Index of Mortality Score; IQR= Interquartile range
138
Outcome variables according to mechanical ventilation episodes at prospective study
Table 7.2 shows the outcome variables according to mechanical ventilation episodes. The median
duration of mechanical ventilation was 4.9 days (IQR: 2.9 - 8.0), median length of PICU stay 7.7
days (IQR: 5.0 - 14.7), and median length of stay of hospital was 20.6 days (IQR: 10.3 - 52.4). There
were 12 patients still in PICU, and five deaths reported when data collection ended on the 12 th
December 2017. Outcome variables were homogenous between prospective and retrospective study
(p>0.05).
Table 7.2: Outcome variables according to mechanical ventilation episodes prospective versus
retrospective study (n=120; n= 262)
Variables Prospective
n (%)
Retrospective
n (%)
p-value
Duration of
mechanical
ventilation
(days, median
(IQR))
4.9 (2.9- 8.0) 4.2 (2.8- 7.5) 0.097
Length of
PICU stay
(days, median
(IQR))
7.7 (5.0- 14.7) 7.6 (4.8- 13.5) 0.32
Length of
hospital stay
(days, median
(IQR))
20.6 (10.3- 52.4) 18.9 (9.8- 34.4) 0.32
PICU
outcome
0.25
Discharge to
ward/home
99 (82.5) 227
Died 5 (4.2) 26
Transferred to
another ICU
(includes
Neonatal ICU)
4 (3.3) 9
Still in ICU 12 (10.0) -
Mortality
(*n=110; n=
234)
0.59
Yes 5 (4.6) 25
No 105 (95.4) 209
* patient level
139
Incidence of VAP and VAE at prospective study
In the prospective study, five episodes of VAP (4.2%) (five patients) and six episodes of VAE (5.0%)
(six patients) were identified in the 120 cases of mechanical ventilation studied (110 patients) using
the PNU1/VAP and the VAE surveillance tools respectively.
The incidence rates for VAP and VAE are given in Table 7.3. As can be seen, there was a reduction
of the VAP rate by 3.9 per 1000 ventilator days (end of ventilation) and 4.4 per 1000 ventilator days
(until the patients were no longer at risk). The rate of VAE dropped by 2.8 per 1000 ventilator days
(end of ventilation) and 3.2 per 1000 ventilator days (until the patients were no longer at risk)
respectively. The different VAP and VAE rates between two phases was not significant (p>0.05).
Table 7.3: Comparison of VAP and VAE incidence between retrospective and prospective studies
measured on end of ventilation and until the patient is no longer at risk
Incidence Retrospective
study
Prospective study Difference/Reduction p-value
(95% CI)
VAP
1) End of
ventilation
9.3 per 1000
ventilator days
5.4 per 1000
ventilator days
3.9 per 1000 ventilator
days
0.31 (-3.61-
11.41)
2) Until the
patient is no
longer at risk
10.2 per 1000
ventilator days
5.8 per 1000
ventilator days
4.4 per 1000 ventilator
days
0.27 (-3.44-
12.24)
VAE
1) End of
ventilation
9.3 per 1000
ventilator days
6.5 per 1000
ventilator days
2.8 per 1000 ventilator
days
0.48 (-4.99-
10.59)
2) Until the
patient is no
longer at risk
10.4 per 1000
ventilator days
7.2 per 1000
ventilator days
3.2 per 1000 ventilator
days
0.45 (-5.02-
11.42)
Of six VAE cases, one was categorised as IVAC, and no mechanical ventilation episode/patient met
the PVAP tier (Table 7.4). One patient met both surveillance tools (VAC as per VAE surveillance
tool and VAP as per PNU1/VAP).
Parallel to the retrospective study, one patient on the same mechanical ventilation episode was
identified as both VAP and VAE, but according to VAE surveillance tool it was further classified as
VAC instead of VAP defined by PNU1/VAP in the prospective study. One patient was further
identified as IVAC instead of VAP defined by PNU1/VAP in the retrospective study.
140
Table 7.4: VAP and VAE counts (tiers) according to mechanical ventilation episode prospective and
retrospective studies
Variable Prospective
n=120
Retrospective
n=262
VAP
Yes 5 16
VAE
VAC 6 16
IVAC 1 4
PVAP 0 3
Discussion
Discussion of demographic and outcome variables of the prospective study
It was found that male patients dominated PICU admissions in the prospective study. Patients aged
less than one year constituted more than half of PICU admissions. In most paediatric VAP research,
this distribution is similar; e.g. Gautam et al. (2012) prospective study undertaken in an Australian
PICU. The PICU and hospital length of stay were similar to those of other paediatric studies, which
report between five and ten days (Bigham et al., 2009; Elward et al., 2002; Gautam et al., 2012; Gupta
et al., 2015; Srinivasan et al., 2009).
Most patient admissions to PICU in the prospective study were by direct admission (44.3%) or via
operation theatre/recovery (34.8%), which is congruent with data from previous years (2014–2016)
reported by the Australian and New Zealand Intensive Care Society, (2017). The PIMS3 score was
4.92%, which was higher than that of the retrospective data; however, this result may not reliably
reflect the final PIMS3 score because this score is only available after patients have been
discharged/died, and some of the patients in this study were still in the PICU/hospital when data
collection ended.
Therefore, the demographic and outcome variables were homogenous between prospective and
retrospective study except for PIMS3. This suggests that the distribution of patients’ demographic
characteristics in QCH’s PICU were comparable throughout the year, allowing meaningful statistical
comparison to assess possible relationships concerning VAP/VAE development.
Discussion of incidence of VAP and VAE at prospective study
VAP rates were measured at 5.4 per 1000 ventilator days and VAE rates were 6.5 per 1000 ventilator
days during the six-month prospective study. These results appear relatively high compared to a
previous study by Narayanan et al. (2016); their six-month prospective study found rates of 2.4 (VAP)
141
and 4.2 (VAE) per 1000 ventilator days. To date, Narayanan’s study has been the only prospective
study to follow paediatric patients for a period of six months, evaluating the application of new VAE
surveillance tools and comparing them to existing PNU1/VAP surveillance tools.
The findings of the present study should be cautiously compared to Narayanan’s study as no
comparable Australian study has been undertaken. One Australian study by Gautam et al. (2012)
evaluated only VAP rates in a one-year prospective study, finding a VAP rate of 7.02 per 1000
ventilator days. The present study found lower VAP rates compared to a recent six-month prospective
study in 16 PICUs in the USA, which found 7.1 per 1000 ventilator days (Gupta et al., 2014). These
differences may be due to the demographic variations in respective PICUs, institutional preventative
strategies, and varying treatment options (Awasthi et al., 2013; Bassant Salah, Sally, & Seham Awad
El, 2017; Srinivasan et al., 2009).
There was a reduction of VAP and VAE in comparison to Phase 1: Retrospective study. The incidence
rate of VAP was reduced by 3.9 per 1000 ventilator days and the VAE incidence rate to 2.8 per 1000
ventilator days. These reductions are within the expected range seen in other published VAP studies
assessing the changes following implementation of VAP education, compliance auditing with
feedback, and VAP bundle implementation in PICUs (Bigham et al., 2009; Brierley et al., 2012;
Casado et al., 2011; Gupta et al., 2014; Rosenthal, Alvarez-Moreno, et al., 2012). These reductions
did not reach statistical significance; however, similar findings were reported in Gupta et al. (2014),
where the incidence of VAP was reduced by 5.6 pre to post implementation of VAP education (20.2
versus 14.6 per 1000 ventilator days respectively) with p=0.21.
Before this present study was undertaken, there had been no specific paediatric study undertaken to
prospectively evaluate VAE rates before and after education, compliance auditing and feedback. The
reduction rate of VAE found in this study was within the ranges found by Cocoros et al. (2016) in
their multi-centred paediatric retrospective study (3.3 to 4.6 per 1000 ventilator days), but they did
not assess the VAE rate after any interventions.
A possible explanation for the reduction in VAP/VAE rates is the ability for daily surveillance to
detect early changes due to nosocomial infections (Ford-Jones et al., 1989). Another possible reason
behind this reduction could be the positive improvement derived from VAP education, compliance
auditing and feedback, as well as other quality improvement activities commenced concurrently in
the study setting (see Chapter 4, Section 4.4). Despite statistical insignificance, clinically, this study
142
suggests a positive influence of VAP education and compliance auditing with feedback undertaken
in unit. The relationship between preventative strategies and the reduction of VAP rates has been
previously seen in PICUs where similar approaches have been implemented (Bigham et al., 2009;
Brierley et al., 2012; De Cristofano et al., 2016; Gupta et al., 2014; Obeid et al., 2014).
Overall, in this study the VAE rates were the same or higher than VAP rates when the two surveillance
tools were applied in both the retrospective and prospective studies. In the retrospective and
prospective studies one patient met both VAP and VAE (using both surveillance tools respectively)
but appeared different in terms of further classification based on VAE tier; IVAC and VAC versus
VAP using PNU1/VAP. This suggested that the new VAE tool described complications other than
VAP or the criteria in respective tiers (VAC, IVAC and PVAP) are strict enough to distinguish
whether or not there is VAP in mechanically ventilated children. These results further support the use
of the VAE surveillance tool in extending beyond VAP identification as seen in previous adult studies
and a few paediatric studies (Cirulis et al., 2016; Klompas, 2013b; Lutmer & Brilli, 2016; Magill et
al., 2014).
Potential risk factors according to mechanical ventilation episodes
Sixty three percent of mechanical ventilation episodes were a single episode of intubation. Some
patients in this cohort were administered paralytic agents (31.7%) and gastrointestinal (GI)
prophylaxis (61.7%). Just over half (53.8%) of the patients received mechanical ventilation via nasal
endotracheal tubes (ETT), and most (74.2%) received light sedation during mechanical ventilation.
All patients had a nasogastric tube in-situ. The majority were intubated with a cuffed ETT (96.7%).
Only 43.3% of patients in this cohort received steroids while 57.5% received blood products at some
time during mechanical ventilation.
No data were collected for ETT cuffed tube, steroid prescription or blood transfusion in the
retrospective study. A homogeneity test for applicable risk factors between the prospective and
retrospective studies revealed that these risk factors were not comparable to each other (p<0.05)
(Table 7.5).
143
Table 7.5: Potential risk factors according to mechanical ventilation episode (n= 120) in prospective
study, compared to retrospective study (n= 262)
Risk Factors Prospective
n (%)
Retrospective
n (%)
p-value
Reintubation 0.033
Yes 44 (36.7) 68 (26.0)
No 76 (63.3) 194 (74.0)
Paralytic agent 0.030
Yes 38 (31.7) 56 (21.4)
No 82 (68.3) 206 (78.6)
GI prophylaxis 0.006
Yes 74 (61.7) 122 (46.6)
No 46 (38.3) 140 (53.4)
Nasogastric presence n/a
Yes 120 (100) 259 (98.9)
No 0 (0) 3 (1.1)
Routes of intubation 0.038
Nasal 64 (53.4) 169 (64.5)
Oral 56 (46.6) 93 (35.5)
Sedation level
Deep sedation 31 (25.8) 15 (5.7) 0.0001
Light sedation 89 (74.2)
247 (94.3)
ETT types Cuffed 116 (96.7) - n/a
Un-cuffed 4 (3.3)
Steroid prescriptions Yes 52 (43.3) - n/a
No 68 (56.7)
Blood transfusion Yes 69 (57.5) - n/a
No 51 (42.5)
ETT=Endotracheal tube; n/a=not applicable
Compliance of preventative strategies of VAP in the prospective study
Table 7.6 shows the compliance rates for VAP preventative strategies implemented in the prospective
study. The overall compliance to VAP preventative strategies in this phase was 90.9%. Hand hygiene
compliance in the prospective study was sustained within the Australian Hand Hygiene Initiative
benchmark and congruent to the retrospective study. Likewise, in the retrospective study, ETT
suctioning achieved full compliance.
Oral hygiene was performed regularly, averaging five times/day; this was close to the unit’s guideline
(six times/day), achieving a compliance rate of 88.3%. In addition, the majority of PICU staff adhered
to 12-hourly oral health assessments (94.2%) and adhered to oral hygiene based on age-appropriate
factors stated in the PICU guideline (65.0%). The median frequency for ETT cuff pressure assessment
144
was 2.5 times/day which achieved the recommended compliance of twice per day. Nurses
demonstrated 90.0% adherence to 12-hourly cuff pressure checks and consistently maintained
pressure within the range limits set by the unit.
Early commencement of enteral feeding within 24 hours of PICU admission achieved full
compliance. This study revealed that 65.0% of patients with no contraindication had received enteral
feeding within 24 hours of admission, while the remaining 35.0% had reasonable clinical indications
which prevented them from receiving feeding.
Overall, the compliance rate for VAP preventative strategies in the prospective study showed an
increase of 1.9% from the retrospective study (89.0%). Hand hygiene compliance was maintained
within national benchmarks in both phases of the study. Descriptively, other individual preventative
strategies were recorded as having increases in compliance.
145
Table 7.6: Comparison of VAP compliance preventative strategy performance with PICU standard/National standard
VAP Preventative Strategies Mean (SD)
n (%) PICU/ national standard % compliance in comparison
to PICU /national standard
Hand hygiene 84.1% (3.7) > 80% ≥100
Oral hygiene 5.3 (1.1) 6 times/24 hours 88.3
Adhered to 12-hourly oral
health assessment
Yes 113 (94.2) 2 times/24 hours 94.2
No 7 (5.8)
Adhered to age-appropriate oral
hygiene guideline
Yes 78 (65.0) i) Child under 6 months old without teeth
Moistening (4 hourly or 6 times)
Pink swab with sterile water (0800 & 2000; 1200 & 2400;
0400 & 1600)
ii) Above 6 months with teeth
Tooth brush (2 times) - Toothpaste (0800 & 2000)
Mouth rinse (2 times) - Chlorhexidine (1200 & 2400)
Moistening (2 times) - Pink swab with sterile water (0400
& 1600)
65.0
No 42 (35.0)
ETT suctioning 7.0 (2.2) 4 times/ 24 hours ≥100
HOB elevation (median (IQR)) 22.8 (21.3-
23.8)
24 times/24 hours 95.0
Cuff pressure checks (median
(IQR)) (*n=114)
2.5
(2.0 -3.7)
2 times/24 hours ≥100
Adhered to 12 hourly cuff
pressure checks
Yes
102 (89.5)
89.5
No 12 (10.5)
146
Maintained cuff pressure
readings within the limit
Yes 108 (94.7) Min 10cmH2O to max 20cmH2O 94.7
No 6 (5.3)
Ventilator circuits checks 18.7 (5.3) 24 times/24 hours 77.9
Enteral feeding commencement
within 24 hours of PICU
admission
Yes 78 (65.0) Started within 24 hours of admission if not contraindicated 100
No 42 (35.0)
(contra-
indicated)
SD=Standard deviation; IQR= Interquartile range
147
Compliance of VAP preventative strategies: Retrospective versus prospective studies
The overall compliance to VAP preventative strategies in the prospective study increased from 89.0%
to 90.9%. As shown in Figure 7.1, the univariate analyses revealed the difference between
retrospective vs. prospective data in relation to the frequency of performance for oral hygiene, HOB
elevation, ETT cuff pressure checks, and ventilator circuit checks (p<0.05). Early enteral feeding is
presented in the figure but is not comparable to the retrospective study because it was not compared.
Figure 7:1: Comparison of individual VAP preventative strategy compliance — retrospective versus
prospective studies; p-values given where change is statistically significant.
Discussion of VAP preventative strategies compliance — retrospective versus
prospective studies
The increment of overall compliance for VAP preventative strategies as found in the prospective
study versus the retrospective study mimics the findings of several other VAP studies (Al-Thaqafy,
El-Saed, Arabi, & Balkhy, 2014; Babcock et al., 2004; Bigham et al., 2009; Bouadma, Mourvillier,
Deiler, Le Corre, et al., 2010; Brierley et al., 2012; Lawrence & Fulbrook, 2012). Interestingly,
differences between individual VAP preventative strategies reported in the retrospective versus the
prospective study were statistically significant (oral hygiene, HOB elevation, cuff pressure and
ventilator checks). The statistically significant compliance improvement between prospective and
retrospective study found in this study, however, may not strongly suggest improvement at practise
100
81.6
10091.6 90
70.8
100
88.3
10095
100
77.9
100
0
20
40
60
80
100
120
Hand hygiene Oral hygiene
(p= 0.009)
ETT
suctioning
HOB
elevation (p=
0.014)
Cuff pressure
checks (p=
0.001)
Ventilator
circuits
checks (p=
0.025)
Early enteral
feeding
% o
f co
mp
lian
ce
VAP preventative strategies
Retrospective Prospective
148
level. This may be due to challenges in sustaining compliance in behavioural change (Bigham et al.,
2009; Bouadma, Mourvillier, Deiler, Le Corre, et al., 2010). For example, a recent study by Belela-
Anacleto, Kusahara, Peterlini, and Pedreira (2019) revealed that social pressure of PICU staff
influenced the intention to perform hand hygiene. Minor improvement in hand hygiene compliance
was reported after infrastructure, educational and performance feedback interventions were
implemented.
The results of this study may be explained by several factors; updated VAP education and compliance
auditing with feedback as undertaken may have resulted in compliance improvement. Updated VAP
preventative strategies were made available in the TEACHQ platform from August 2016. During the
compliance auditing, reinforcement for PICU staff to visit the online module was emphasised and
staff were encouraged to view the displayed posters on VAP bundles. These served as a reminder to
PICU staff to practise and maintain compliance with VAP preventative strategies (Borgert et al.,
2015; Flodgren et al., 2013; Hamishehkar et al., 2014; Jansson et al., 2013). The differences may also
have been associated with active compliance auditing and feedback which were concurrently
delivered during and after compliance auditing. Many studies have shown that compliance
enforcement coupled with feedback are key strategies to minimising non-compliance with hospital
guidelines (Helder et al., 2010; Klompas, Branson, et al., 2014; Kollef, 2011; Rosenthal, Guzman,
Pezzotto, & Crnich, 2003) .
Hand hygiene and ETT suctioning remained at full compliance rates in the prospective study, likely
due to ongoing and routine hand hygiene compliance auditing by the hospital staff. This strategy may
encourage long-term compliance, as demonstrated across a number of studies (Bouadma, Mourvillier,
Deiler, Derennes, et al., 2010; Bouadma, Mourvillier, Deiler, Le Corre, et al., 2010; Rosenthal,
Guzman, & Safdar, 2005). With ETT suctioning, the rate of compliance may be due to the nature of
the procedure itself in which the nurse assesses the clinical indications prior to performing it, and this
likely serves as a reminder to perform the preventative strategy (Davies et al., 2011). Enteral feeding
commencement had 100% compliance. This result is to be expected, as in this study setting it is
assessed on admission and at regular intervals during patient rounding thereafter (Bouadma,
Mourvillier, Deiler, Le Corre, et al., 2010).
Univariate analysis
Univariate analyses were conducted to assess associations between possible explanatory variables
and the outcome of either VAP or VAE in the six-month prospective study. A p-value was significant
at ≤0.05.
149
Association of demographic characteristics with VAP and VAE in the prospective study
Univariate analyses for the demographics revealed that none were associated with the development
of VAP and VAE (p-value >0.05; Table 7.7). These findings were consistent with the dataset of the
retrospective study.
Table 7.7: Univariate analysis for association of demographic characteristics with VAP and VAE
(n=120)
Variables VAP p-
value
VAE p-
value
Yes No Yes No
Gender Male 2 (40.0%) 73
(63.5%)
0.36 4 (66.7%) 71 (62.3%) 1.00
Female 3 (60.0%) 42
(36.5%)
2 (33.3%) 43 (37.7%)
Age <1 year 2 (40.0%) 72
(62.6%)
0.28 1 (16.7%) 73 (64.0%) 0.062
1- 12 year 3 (60.0%) 32
(27.8%)
4 (66.7%) 31 (27.2%)
13 and above 0 (0.0%) 11 (9.6%)
1 (16.7%) 10 (8.8%)
Weight 15.8 (5.2 -
29.0)
5.7 (3.5-
15.0)
0.29 23.1 (7.6-
48.5)
5.5 (3.5-
15.2)
0.084
PICU
Diagnosis
Category
Medical 4 (80.0%) 69
(60.0%)
0.65 12 (75.0%) 160
(65.0%)
0.21
Surgical 1 (20.0%) 46
(40.0%)
4 (66.7%) 43 (37.7%)
PICU
source of
admission
OT/ Recovery 1 (20.0%) 42
(36.5%)
0.27 4 (66.7%) 39 (34.2%) 0.24
Emergency Department 0 (0.0%) 9 (7.8%) 1 (16.7%) 8 (7.0%)
Ward (other inpatient
area)/other
ICU/Neonatal ICU-same
hospital
2 (40.0%) 13
(11.3%)
0 (0.0%) 15(13.2%)
Direct admission 2 (40.0%) 51
(44.3%)
1 (16.7%) 52(42.6%)
Underlying
disease
Cardiovascular 0 (0.0%) 33
(28.7%)
0.36 0 (0.0%) 33(28.7%) 0.20
Respiratory 2 (40.0%) 29
(25.2%)
3 (50.0%) 28(24.6%)
Others 3 (60.0%) 53
(46.1%)
3 (50.0%) 53(46.5%)
PIMS3
score
PIMS3 risk for death 1.42(0.66-
28.14)
1.71(0.51-
3.44)
0.81 0.95(0.33-
2.58)
1.71 (0.53-
3.59)
0.25
150
Association of patient outcome characteristics with VAP and VAE
Table 7.8 shows that duration of mechanical ventilation and length of PICU are associated with VAP
and VAE occurrences (p<0.05). This was also found in the retrospective study, but not for the length
of hospital stay (p>0.05).
Table 7.8: Univariate analysis for association of outcome characteristics with VAP and VAE (n=120)
Variables VAP p-
value
VAE p-
value
Yes No Yes No
PICU
outcomes
Discharge
ward/home &
Trans to
another
ICU/Neonatal
ICU
4(80.0%) 96 (83.5%) 0.65 5(83.3%) 95(83.3%) 0.45
Died in ICU 0 (0.0%) 8 (7.0%) 1 (16.7%) 7 (6.1%)
Still in ICU 1 (20.0%) 11 (9.6%) 0 (0.0%) 12 (10.5%)
Mortality
Died
0 (0.0%)
8 (7.0%)
1.00
5 (83.3%)
107 (93.9%)
0.35
Not died 5 (100.0%) 107
(93.0%)
1 (16.7%) 7(6.1%)
Duration of
mechanical
ventilation
(days, median
(IQR))
10.5 (8.8-
6.4)
4.8 (2.9-
7.2)
0.001 19.8 (10.8-
29.2)
4.7 (2.9-6.9) 0.001
Length of
PICU stay
(days median
(IQR))
18.9 (12.1-
89.9)
7.6 (4.9-
13.9)
0.013 22.4 (13.5-
56.7)
7.5 (4.9-13.7) 0.004
Length of
hospital stay
(days median
(IQR))
47.1 (22.2-
70.8)
19.7 (10.2-
42.6)
0.13 52.6 (19.0-
99.5)
18.7 (10.1-
43.8)
0.082
IQR=Interquartile range
Association of possible risk factors with VAP and VAE
VAP was not found to be associated with any of the possible risk factors examined (p>0.05). The
presence of GI prophylaxis and ETT types (cuffed) were associated with the development of VAE
(p<0.05; Table 7.9) in the prospective study.
151
Similarly, the presence of GI prophylaxis was associated in the development of VAE at univariate
analysis in the retrospective study. In contrast to data in the retrospective study, no significant
association was found for reintubation and development of VAE in the prospective dataset.
Table 7.9: Univariate analysis for association of possible risk factors with VAP and VAE (n=120)
Variables VAP p-
value
VAE p-
value
Yes No Yes No
Reintubation Yes 4 (80.0%) 40 (34.8%) 0.060 3 (50.0%) 41 (36.0%) 0.67
No 1 (20.0%) 75 (65.2%) 3 (50.0%) 73 (64.0%)
Paralytic agent Yes 2 (40.0%) 36 (31.3%) 0.65 1 (16.7%) 37 (32.5%) 0.66
No 3 (60.0%) 79 (68.7%) 5 (83.3%) 77 (67.5%)
GI prophylaxis Yes 3 (60.0%) 43 (37.4%) 0.37 5 (83.3%) 41 (36.0%) 0.030
No 2 (40.0%) 72 (62.6%) 1 (16.7%) 73 (64.0%)
Nasogastric
presence
Yes 5 (100.0%) 115 (100.0%) - 6 (100.0%) 114 (100.0%) n/a
No 0 (0%) 0 (0%) 0 (0%) 0 (0%)
Routes of
intubation
Nasal 3 (60.0%) 61 (53.0%) 1.00 1 (16.7%) 63 (55.3%) 0.096
Oral 2 (40.0%) 54 (47.0%) 5 (83.3%) 51 (44.7%)
Sedation level Deep
sedation
1 (20.0%) 30 (26.1%) 1.00 3 (50.0%) 28 (24.6%) 0.18
Light
sedation
4 (80.0%) 85 (73.9%) 3 (50.0%) 86 (75.4%)
ETT types Cuffed 4 (80.0%) 112 (97.4%) 0.16 4 (66.7%) 112 (98.2%) 0.012
Un-
cuffed
1 (20.0%) 3 (2.6%) 2 (33.3%) 2 (1.8%)
Steroid
prescriptions
Yes 4 (80.0%) 48 (41.7%) 0.17 4 (66.7%) 48 (42.1%) 0.40
No 1(20.0%) 67(58.3%) 2(33.3%) 66(57.9%)
Blood transfusion Yes 4(80.0%) 65(56.5%) 0.39 5(83.3%) 64(56.1%) 0.24
No 1 20.0%) 50(43.5%) 1(16.7%) 50(43.9%)
Association of preventative strategies with VAP and VAE
In the prospective study, the univariate analysis revealed that only ETT suctioning was associated
with VAP (p-value <0.05; Table 7.10). No other preventative strategies were found to be associated
with VAE development. In contrast, only cuff pressure checks were associated with the VAP and
VAE in retrospective study (p ≤ 0.05).
152
Table 7.10: Univariate analysis for association of VAP preventative strategies with VAP and VAE
(n=120)
Variables VAP p-
value
VAE p-
value
Yes No Yes No
Hand hygiene 84.4 (4.9) 84.1 (3.6) 0.87 85.7 (3.6) 84.1 (3.7) 0.29
Oral hygiene 5.1 (1.1) 5.3 (1.1) 0.62 5.0 (0.9) 5.3 (1.1) 0.56
a) Adhered to 12-hourly
oral health assessment
Yes 5 (100.0%) 108 (93.9%) 1.00 5 (83.3%) 108 (94.7%) 0.31
No 0 (0.0%) 7 (6.1%) 1 (16.7%) 6 (5.3%)
b) Adhered to age-
appropriate oral hygiene
guideline
Yes 1 (20.0%) 4 (80.0%) 0.050 2 (33.3%) 76 (66.7%) 0.18
No 77 (67.0%) 38 (33.0%) 4 (66.7%) 38 (33.3%)
Endotracheal
suctioning
9.9 (3.5) 6.9 (2.0) 0.002 7.5 (2.9) 6.9 (2.1) 0.59
Head of bed elevation
(median (IQR))
22.6 (22.2- 23.8) 22.8 (21.3-
23.8)
0.69 21.9 (21.0-
22.7)
22.9 (21.3-
23.8)
0.24
Cuff pressure checks
(median (IQR)) (*n=
115)
2.8 (2.0- 3.4) 2.5 (2.0-3.7) 0.92 2.8 (1.0-
3.6)
2.5 (2.0- 3.7) 0.59
a) Adhered to 12 hourly
cuff pressure checks
Yes 4 (100.0%) 98 (89.1%) 1.00 3 (75.0%) 99 (90.0%) 0.36
No 0 (0.0%) 12 (10.9%) 1 (25.0%) 11 (10.0%)
b) Maintained cuff
pressure readings within
the limit
Yes 4 (100.0%) 104(94.5%) 1.00 4(100.0%) 104(94.5%) 1.00
No 0 (0.0%) 6 (5.5%) 0 (0.0%) 6 (5.5%)
Ventilator circuits
checks
18.6 (5.3) 18.7 (5.5) 0.59 16.1 (1.74) 18.7 (5.6) 0.25
Enteral feeding
commencement within
24 hours of PICU
admission
Yes 1 (20.0%) 77(67.0%) 0.050 6(100.0%) 72(63.2%) 0.090
No 4 (80.0%) 38(33.0%) 0 (0.0%) 42(36.8%)
Discussion of demographic and outcome characteristics and VAP/VAE — prospective
versus retrospective studies
No significant association was found between the demographic characteristics in the development of
VAP and VAE by univariate analyses. The results for VAP found in these studies were also observed
by Gupta et al. (2015) and Gautam et al. (2012) in their prospective studies. The development of VAP
153
and VAE was found to be directly associated with the duration of mechanical ventilation. Again, this
result for VAP was consistent with previous prospective studies by Gupta et al. (2015) using
univariate analysis, and the duration of mechanical ventilation was found to be an independent risk
factor by multivariate analyses by Casado et al. (2011) and Awasthi et al. (2013). These findings may
further support the idea that patients are more readily exposed to complications which may precipitate
VAP and VAE. Most importantly, VAP/VAE preventative strategies that could assist in reducing the
duration of mechanical ventilation should be prioritised to prevent and minimise chances of
infection/events (Goutier et al., 2014; Klompas, Branson, et al., 2014; Muscedere et al., 2013; Neto
et al., 2015; Sinuff et al., 2013).
Patients developing VAP/VAE in the present study were associated with a longer duration of
mechanical ventilation and PICU stay. These findings were also reported in other VAP and VAE
studies in paediatrics (Bigham et al., 2009; Cirulis et al., 2016; Cocoros et al., 2016; Gautam et al.,
2012; Gupta et al., 2015; Phongjitsiri et al., 2015; Srinivasan et al., 2009). The findings support the
association of ventilator usage and the development of VAP/VAE in PICU.
Discussion of potential risk factors and preventative strategies and VAP/VAE —
prospective versus retrospective studies
The univariate analysis in the prospective study found that patients intubated with a cuffed ETT and
patients with a GI prophylaxis prescription were associated with the development of VAE. These
results have not been found by any parallel VAE studies in children to date. The association between
cuffed ETTs and VAE development in this study could be related to ETT management, particularly
in keeping the ETT cuff adequately inflated (Gupta & Rosen, 2016; Weiss et al., 2009).
Only GI prophylaxis was found to have a significant association with the development of VAE in the
prospective and retrospective studies. This result has not previously been described in paediatric
settings (Beardsley et al., 2016; Cirulis et al., 2016; Cocoros et al., 2016; Phongjitsiri et al., 2015).
An adult study by Klompas et al. (2016) found that GI prophylaxis was associated with the
development of PVAP by using the VAE surveillance tool. Evidence about the underlying mechanism
behind this is still unclear, but this finding may suggest that the undesired effects of GI prophylaxis
may also be present in children (Albert et al., 2016).
In the prospective study, additional potential risk factors for VAP/VAE were evaluated which were
not examined during the retrospective study. These factors include steroid and blood product
administration, but these risk factors did not reveal any statistically significant results in relation to
154
the development of VAP and VAE. Blood transfusions were associated with VAE development in
the bivariate analysis by Cocoros, Priebe, Gray, et al. (2017), who also found an association with
steroid administrations; however, these were not assessed in this study (Cocoros, Priebe, Gray, et al.,
2017). A possible reason why the present study failed to find an association between these factors
may be due to the small sample size and low rates of VAP/VAE found in the data.
The frequency of ETT suctioning at a minimum of four times/day was associated with the
development of VAP (p=0.002). This finding was consistent with another prospective study by
Awasthi et al. (2013); they advised close monitoring of patients for VAP if the patient required ETT
suctioning more than once in six hours. The findings of the present study suggest that, despite full
compliance with ETT suctioning (identified in this phase), gaps in compliance of other elements
related to suctioning such as adherence to aseptic non-touch techniques (ANTT) and the practise of
draining condensate of ventilator circuits were identified. It is noted that, in compliance auditing,
these two sub-elements of ETT suctioning were only at a compliance rate of 74.2% and 22.6%
respectively. These elements may predispose the patient to the development of VAP (Auxiliadora-
Martins et al., 2012; Klompas, Branson, et al., 2014; Resar et al., 2014; Tolentino-DelosReyes et al.,
2007).
Summary
The six-month prospective study revealed a reduction in VAP/VAE incidence rates in comparison to
the retrospective study, measured by the PNU1/VAP and the new VAE surveillance tools. These
results are further supported by statistically significant improvements in VAP preventative strategy
compliance, including oral hygiene, HOB elevation, ETT cuff checks, and ventilator circuit checks
as observed at this phase in comparison to the retrospective study, but cautiously inferred to clinical
practise. Univariate analysis revealed that GI prophylaxis was associated with VAE development
which was identified in both the retrospective and prospective studies, warranting further research
around this risk factor in VAE.
Chapter 8 presents a general discussion for the thesis.
155
General Discussion
Introduction
This chapter provides a summary of ventilator-associated pneumonia (VAP) and ventilator-associated
events (VAE) in children, the challenges faced in preventing these events, and knowledge gaps in the
literature. A brief outline of the methodological approaches used and the findings from the three
phases of the study will be synthesised to draw conclusions. Finally, the chapter will detail the
limitations and implications of this study as well as recommendations for future research in this field.
Summary of the thesis
Complications related to mechanical ventilation have significant impacts on the health of patients.
Invasive mechanical ventilation is delivered with the intention of maintaining gas exchange in a
patient with as few adverse effects as possible (Keszler, 2017). For decades, any complication related
to mechanical ventilation have been viewed purely through a VAP lens (Hayashi et al., 2013;
Klompas, 2012). In the last six to seven years, there has been increased interest in VAP research
amongst medical, nursing and allied health professions (Magill et al., 2013; Mietto et al., 2013). This
interest has stemmed from a paradigm shift in VAP understanding to a broader scope of
complications, such as non-infectious conditions; i.e., atelectasis and pulmonary oedema. This shift
in thinking prompted a review and update of the surveillance tool. The controversy around VAP
surveillance tools continues; however the introduction of the Centre for Disease Control and
Prevention (CDC)’s VAE surveillance tool for adults provides a new foundation for research in
ventilator-associated complications (Klompas, 2013b; Klompas et al., 2015; Stevens et al., 2014). In
paediatrics VAP presents both a clinical and surveillance challenge to overcome. There are also
persistent challenges regarding surveillance tools in adults (Venkatachalam et al., 2011; Wright &
Romano, 2006). The body of evidence informing VAP/VAE prevention strategies is growing.
Currently the epidemiological data in paediatrics is less robust than that for adults, which may restrict
prevention strategy development (Aelami et al., 2014; Srinivasan et al., 2009).
The literature presented in Chapter 2 showed the varied incidence of VAP across PICU settings,
including the dominance of modifiable risk factors over non-modifiable risk factors (Awasthi et al.,
2013; Bigham et al., 2009; Cocoros, Priebe, Gray, et al., 2017; Gautam et al., 2012; Gupta et al.,
2015; Kusahara et al., 2014; Roeleveld et al., 2011).Very few studies have been undertaken using the
VAE surveillance in children. In particular, evidence of the application of VAE surveillance tools in
the paediatric population is currently limited to the existing VAP surveillance tool and does not extend
to cover VAE (Mohd Ali, Jauncey-Cooke, & Bogossian, 2019).
156
The literature appraised in Chapter 3 reported on individual VAP preventative strategies as currently
applied to children. Evidence exploring the individual preventative strategies in the bundle was
appraised. Individual preventative strategy compliance measurements were found to be unclear and
varied throughout the reviewed studies, and some did not report the compliance benchmarks to which
they were adhering (Tobias et al., 2012). The overall compliance benchmark for VAP preventative
strategies for adults is capped at >95% by the Institute for Healthcare Improvement (IHI) (Resar et
al., 2014), but some individual preventative strategies are not specified. Furthermore, some of these
preventative strategies used in the adult bundle were not applicable, or the effects were unclear in
children. Thus, translation of adult preventative strategies evidence (such as that concerning GI
prophylaxis and sedation interruption) to paediatrics poses uncertainty as to whether these would
assist VAP prevention in children (Klompas, Branson, et al., 2014; Lopriore et al., 2002; Vet et al.,
2016; Yildizdas et al., 2002). Minimal data on VAE preventative strategies is currently available
(Cocoros et al., 2016; Cocoros, Priebe, Gray, et al., 2017; Phongjitsiri et al., 2015) as a result of lack
of study in paediatric population.
Evidence continues to support ongoing challenges in VAP preventative strategy implementation in
the clinical setting, with issues such as a lack of compliance among healthcare workers (Bigham et
al., 2009; Brierley et al., 2012; Nair & Niederman, 2015; Rello et al., 2013; Smiddy et al., 2015). An
education program has been found to be beneficial to patient safety by maximising adherence to the
latest evidence-based practise recommendations (Flodgren et al., 2013; Jansson et al., 2013; Siegel et
al., 2007). However, a review of the literature identified that VAP education in PICUs is under
researched, and the possible benefits of parental involvement in VAP strategies is an opportune area
which has not been explored (Bellissimo-Rodrigues et al., 2016; Ciofi degli Atti et al., 2011). The
initiative ‘Speaking up for hand hygiene’ revealed a promising partnership between parents and PICU
staff, although the literature revealed mixed perceptions from those involved (Bsharat & Drach-
Zahavy, 2017; Pan et al., 2013; Wu et al., 2013). Methods to maximise this partnership require further
investigation. These identified gaps in knowledge informed the three phases of study design detailed
in Chapter 4; Phase 1: Retrospective study; Phase 2: VAP education, VAP preventative strategies
compliance auditing and surveys; and Phase 3: Prospective study.
Phase 1: The retrospective study provided the latest VAP/VAE epidemiological data reported from
one of the largest paediatric hospitals in Australia (Queensland Children’s Hospital (QCH)) in the
calendar year 2015. The incidence of VAP found in this study was higher compared to recent
prospective studies conducted in single PICUs in Sydney by Gautam et al. (2012) and was the first
study specifically focussed on surveillance for VAE incidence rates in an Australian paediatric ICU.
157
In the present study, the VAE incidence rate was lower than those of two paediatric studies which
adapted the VAE surveillance method for adults to children (Iosifidis et al., 2016; Phongjitsiri et al.,
2015). The rates were higher than those studies which applied some modification to adult VAE
surveillance tools — using mean arterial pressure (MAP) instead of positive end expiratory pressure
(PEEP) and applied PEEP of 2cmH2O instead of 3cmH2O (Beardsley et al., 2016; Cocoros et al.,
2016). The VAE surveillance tool appears to be stricter, helping isolate complications related to
mechanical ventilation. Three mechanical ventilation episodes met the PVAP versus 16 mechanical
ventilation episodes which met the PNU1/VAP tool. In this case VAE surveillance tool in paediatric
had classified 13 mechanical ventilator episodes as other possible ventilator-associated
complications.
The overall compliance of VAP preventative strategies was 89.2% in the retrospective study.
Multivariate analysis findings suggest that increased frequency of oral hygiene performance and GI
prophylaxis prescriptions may cause more harm to paediatric patients, which was congruent with an
adult study by Klompas et al. (2016). Thus, higher levels of evidence from randomised controlled
trials are required for future research to confirm these findings.
Given the ongoing challenges with VAP preventative strategy implementation in PICUs, VAP
education and compliance auditing with feedback were carried out in Phase 2. These approaches are
consistent with the most frequently used implementation of care bundles in ICUs (Borgert et al.,
2015). In this study, the involvement of parents and staff was increased including assessing their
perceptions in the ‘Speaking up for hand hygiene’ initiative. The overall compliance rate for VAP
prevention strategies in this phase was within an acceptable range (Tabaeian et al., 2017). While hand
hygiene compliance among PICU staff was in the range of the Australian national benchmark (>80%),
parents’ hand hygiene compliance was reported at 64.7%. This finding provides insight into the
importance of hand hygiene education and compliance monitoring among parents. Similarly, some
individual VAP preventative strategies/sub-elements such as oral hygiene performance, 12-hourly
cuff pressure checks, and adherence to aseptic non-touch techniques during ETT suctioning were
reported at compliance within a relatively acceptable range (50–75%) (Tabaeian et al., 2017). These
findings suggest that there may be some potential for further improvement, and therefore consistent
compliance auditing may help to close this gap. The findings from the parent and nurse surveys
supported the benefits derived from ‘Speaking up for hand hygiene’ in PICUs. Overall, there was a
positive perception of the willingness to remind each other to perform hand hygiene. These findings
suggest a future study with a larger sample size of parents and nurses would be required to reach a
158
firm conclusion. The reasons reported by parents and nurses that make them reluctant to remind each
other to perform hand hygiene warrant further exploration.
To evaluate the changes made in the PICU after delivering updated VAP education, and compliance
auditing with feedback and surveys (see Chapter 4, Section 4.4.3), a six-month prospective study was
conducted where the incidence rates of VAP/VAE were enumerated. The results showed that the rates
were reduced and there was an increase in overall compliance with preventative strategies. This
improvement of overall compliance was supported by the statistically significant difference reported
for individual VAP preventative strategies for oral hygiene, HOB elevation, ETT cuff pressure
checks, and ventilator circuit checks. Clinically, the improvement of compliance is often associated
with improved incidence rates of VAP. These results supported the idea that updated VAP education
for nurses and parents, and compliance auditing with feedback were able to significantly increase
compliance rates, and this was associated with a reduction in VAP/VAE incidence in the PICU. Due
to low event rates in this study, the analysis of possible risk factors and VAP preventative strategies
were limited to univariate analysis only.
New VAE surveillance tool for global surveillance
VAP surveillance activities are crucial for internal quality assessment and external benchmarking in
intensive care settings (Klompas, Branson, et al., 2014; Klompas, Kleinman, et al., 2012; Rosenthal,
Alvarez-Moreno, et al., 2012). The impact of VAP relates not only to morbidity, but also to increases
in healthcare expenditure (Muscedere, Day, & Heyland, 2010). Despite the limitations of current
VAP surveillance tools, surveillance is vital to assess VAP incidence rates, the development of
preventative strategies and strategic planning (Hebert et al., 2017; VICNISS Healthcare Associated
Infection Surveillance, 2015). Furthermore, surveillance enables evaluation of improvements in VAP
preventative measures undertaken in the unit (Wong, Mathieu, & Williams, 2012).
Lapses in surveillance can have a major impact. Benet, Allaouchiche, Argaud, and Vanhems (2012)
evaluated the impact of a lapse of one year in VAP surveillance that resulted in an increased VAP
rate from 13.4 to 22.9 per 1000 ventilator days and increased the duration of mechanical ventilation
from 7.7 to 11.3 days (p= 0.007). In the USA, surveillance data reporting to the National Healthcare
Safety Network (NHSN) is mandatory for healthcare facilities as an inpatient quality indicator and
for hospital reimbursement purposes (Magill et al., 2013; Stoeppel et al., 2014). However, in Australia
and New Zealand VAP surveillance is an optional module at a national level (VICNISS Healthcare
Associated Infection Surveillance, 2015). The barriers to mandatory VAP surveillance include the
159
substantial time required, difficulties in applying the surveillance criteria, and often a lack of
confidence at an institutional level regarding what to do with the data obtained from surveillance
(Friedman et al., 2005).
At present globally, PNU/VAP surveillance is considered adequate to monitor VAP in children
(Lutmer & Brilli, 2016), but this surveillance proves challenging for hospitals to maintain due to the
subjectivity of its criteria and difficulty in implementation (Hayashi et al., 2013; Klompas, 2010,
2012; Magill et al., 2013; Magill et al., 2014). This study has offered the groundwork to support the
validation of the VAE surveillance tools in paediatrics. This VAE tool is believed to overcome the
limitations of the present PNU/VAP surveillance through its potential to increase validity and
reliability (Boyer et al., 2015; Klompas, 2013a; Phongjitsiri et al., 2015). Initial efforts to test the
feasibility of VAE surveillance in children started in 2012 when the CDC Paediatric and Neonatal
Ventilator-Associated Event Working Group was formed. Subsequent discussion has concluded that
additional data is required before this will be mandated in mechanically ventilated paediatric patients
(Cocoros et al., 2016). VAE surveillance has only recently been tailored for use with children,
although the surveillance tool has been used in adult ICUs (Cocoros et al., 2016; Cocoros, Priebe,
Logan, et al., 2017; Klompas, 2012, 2013a, 2013b). More studies are required to further refine the
current tool so that it is appropriate for PICU use.
The available studies in children to date have involved the evaluation of the current PNU/VAP
surveillance tool and the new VAE surveillance tool (Iosifidis et al., 2016; Narayanan et al., 2016;
Phongjitsiri et al., 2015). However, given the variability and difference between adults and children,
a group of researchers have started testing different VAC criteria in children, including altering FiO2
thresholds and replacing PEEP with MAP, modifying PEEP levels, altering the tool of period of
stability, and reducing abnormal temperature and WBC count (for IVAC) in the VAE surveillance
tool (Beardsley et al., 2016; Cirulis et al., 2016; Cocoros et al., 2016; Cocoros, Priebe, Logan, et al.,
2017).
Furthermore, the variability of ventilator management in paediatrics, such as in the case of acute
respiratory distress syndrome (ARDS), may also contribute to the significant difference in defining
VAC in children (Khemani, Markovitz, & Curley, 2009; Mhanna, 2016). Further research is needed
to refine the VAE surveillance tool for children in PICUs.
The new VAE surveillance tool has the potential for international comparison for collaborative
surveillance on ventilator-associated complications in children (Cocoros et al., 2016; Cocoros, Priebe,
Logan, et al., 2017; Taylor et al., 2014). However, this is only possible once VAE data becomes
160
readily available globally and consistency in reporting is achieved. A literature review by Mohd Ali
et al. (2019) found that current published VAE data for children is limited to a few countries, with
data only available from five studies in the USA, one study in Greece and one study in the United
Kingdom (Iosifidis et al., 2016; Narayanan et al., 2016).
Lessons learnt from paediatric VAP surveillance include that the incidence rates varied partly due to
the subjectivity of the PNU/VAP surveillance tool and the diversity of demographic characteristics
of the included paediatric patients (Awasthi et al., 2013; Nair & Niederman, 2015; Srinivasan et al.,
2009). Meaningful comparison for planning strategic preventative strategies would benefit from the
use of objective criteria in the VAE surveillance tool and an automatic data extraction device, such
as a VAE dashboard. This may overcome the subjectivity and labour requirements currently
hampering VAP surveillance activities (Hebert et al., 2017).
The role of VAP education, compliance auditing with feedback, and parents’ involvement
in VAP/VAE prevention
This study emphasised the role of VAP education, compliance auditing with feedback, and parental
involvement in the improvement of compliance and the reduction of VAP/VAE rates in a PICU.
Implementation of VAP preventative strategies that addressed staff education and compliance
monitoring with feedback have previously resulted in the reduction of VAP rates significantly, from
52% to 6% in a pre- and post-intervention study (Obeid et al., 2014). Kunzmann, Dimitriades,
Morrow, and Argent (2016) also achieved a VAP reduction, from 55 per 1000 to 19.1 per 1000
ventilator days over the first five months of their study, with greater reductions achieved down to 4
per 1000 ventilator days with an improvement of VAP preventative strategies compliance from 57%
to 83%.
Some researchers have argued that it is unlikely that universally accepted paediatric VAP preventative
measures will be achieved. This is due to the diversity of underlying diseases, pathogenesis of patient
conditions, treatments, and different training and qualifications of PICU staff (Morrow, Argent,
Jeena, & Green, 2009). The VAP education for PICU staff in this study drew on the latest evidence
on individual preventative strategies that best suit paediatric patients. Education on healthcare-
associated infections requires frequent updating to ensure it is in line with epidemiological changes
and advancement in the latest medical technologies (Hellyer et al., 2016; Nair & Niederman, 2015;
Pittet, 2010). These findings are congruent with the findings of a systematic review, which stated that
education strategies through staff education, auditing and feedback, reminders, and VAP surveillance
161
may be associated with an increase in healthcare workers’ knowledge and compliance (Jansson et al.,
2013).
The findings of VAP preventative strategies compliance auditing at Phase 2 showed acceptable
compliance, but some sub-elements of individual preventative strategies were below the acceptable
range. This suggests that these elements should not be ignored and need to be examined thoroughly.
The compliance rates of preventative strategies and the subsequent reduction of VAP/VAE require
sustainable measures in clinical practise as part of a quality improvement framework to address the
complexity of VAP and VAE (Jansson et al., 2013). It is recommended that providing feedback to
healthcare workers on hand hygiene performance, as well as education, will help motivate and
optimise compliance (Yokoe et al., 2014).
This study extends VAP education to parents, encouraging them to contribute to VAP minimisation
in PICU. Parents play an important role in patient care, and this role should therefore be looked at
from a broader perspective. Their voices when speaking up for hand hygiene are one of many ways
to integrate parents’ participation in VAP/VAE prevention in PICUs (Bellissimo-Rodrigues et al.,
2016).
Education should be openly delivered in a simple, informative, and easily understood way, tailored
to the target group to ensure maximum engagement and retention of the lessons within. In this study,
in the QCH PICU environment, pamphlets on hand hygiene promotion and VAP prevention were
given to parents. These pamphlets were intended as a way to maximise parental engagement with
hand hygiene during the time of their child’s PICU admission.
Overall, the present study provides some insight into the need for consistency when updating
educational information on VAP in PICU and involving staff and parents to overcome challenges in
compliance with VAP preventative strategies.
Limitations and strengths
The primary limitation of this study is that it involved a single setting, limiting the scope of advanced
statistical modelling, as the VAP/VAE rates were very low. This ultimately restricted the
generalisability of the data gathered (Al-Thaqafy et al., 2014; Cirulis et al., 2016; Gautam et al.,
2012). The VAP/VAE surveillance tools were applied on the same data, which may potentially have
included inaccurate clinical data. Data collected retrospectively may have been subject to investigator
bias, thus the associations could not be assumed as causal (Hatachi et al., 2015; Phongjitsiri et al.,
162
2015). However, these biases would be equally applicable to both tools. The data from the electronic
records were in the form of real-time data which reduced the possibility of transcription errors, which
may have been encountered if paper-based chart reviews were used instead for data collection (Stein
et al., 2014).
Another restriction of the study lay in the lack of data on the possible risk factors of VAE in children.
When the data collection sheet for the retrospective study was being designed, limited information
on risk factors for VAE in paediatrics had been established (in part due to the lack of information
available in related literature); thus the study aimed to assess the most reported VAP risk factors and
VAP preventative strategies. However, in the prospective study, data on these additional risk factors
and preventative strategies was collected. These additional risk factors add new information to the
literature. Multivariate analysis could not be completed due to the low event rates encountered over
the six-month prospective study (Cocoros et al., 2016). In addition to this, discussion on the
application of the new VAE surveillance tools was limited to a very few paediatric studies, which
only provide scarce evidence to support the possible explanation of what this study has found
(Cocoros et al., 2016; Cocoros, Priebe, Logan, et al., 2017; Phongjitsiri et al., 2015). Thus, the study
should be duplicated in a selection of other PICUs and involve more patients to increase external
validity and the availability of data for analysis and generalisation (Cocoros et al., 2016; Phongjitsiri
et al., 2015).
The survey was undertaken in single study site with a small sample size and low response rates.
Although the parents received the education and pamphlet, in some circumstances they declined to
participate in the survey, contributing to the low response rate. In the nurses’ survey, although the
survey was available online or in a hard copy the response rate remained very low. Several reminders
had no impact on the response rate. This may be due to various reasons such as unit activity or patient
acuity. Response bias is also a study limitation. Parents and nurses may provide the answers to the
surveys that were influenced by the ongoing hand hygiene campaign in the PICU or other source of
information regarding hand hygiene. Thus, caution should be exercised in drawing firm conclusions
based on these findings. Furthermore, the perception of other PICU staff towards parents and nurses
on ‘Speaking up for hand hygiene’ was not examined and remains an area of potential improvement.
Despite the limitations discussed above, this study presents several strengths. The retrospective study
provides robust data on recent VAP epidemiology in an Australian context; surveillance for
VAP/VAE is an optional module with only five adult hospitals in Australia reporting VAE, and no
PICUs or Neonatal ICUs participating in VAP/VAE surveillance (VICNISS Healthcare Associated
163
Infection Surveillance, 2015). Although limited to a single setting and retrospective in nature, this
study is one of the first studies to test the CDC’s VAE surveillance tool in paediatric patients, and
among the first to examine present VAP/VAE preventative strategies in children. Although
insufficient VAP/VAE events were found in this study, two VAP preventative strategies (oral hygiene
and GI prophylaxis associated with VAE) were identified, which may provide a starting point for
further research.
The incidence rate calculation in this study extends the calculation used by the CDC (total number of
days/patients on ventilator) (Centers for Disease Control and Prevention (CDC) & National
Healthcare Safety Network (NHSN), 2015a), to the total time (hours) until the VAP/VAE is
diagnosed, ventilator is ceased in the absence of development of VAP/VAE, or death. In doing so,
the study was able to predict the development of VAP/VAE per hour of ventilation. This is crucial to
be able to revise nursing care or VAP/VAE preventative strategies, especially in terms of frequency
and timing of performance. For instance, the frequency of individual VAP preventative strategies
often described as ‘routine’ failed to distinguish the optimum frequency which would best suit
paediatric patients: the frequency of oral hygiene, endotracheal tube (ETT) cuff pressure checks, ETT
suctioning, and the level of HOB elevation in the practise examples are unclear (Brierley et al., 2012;
Jácomo et al., 2011; Kusahara et al., 2012; Memela & Gopalan, 2014; Morrow & Argent, 2008;
Munro & Ruggiero, 2014; Wip & Napolitano, 2009). Future research should assess in detail the
frequency and timing of VAP preventative strategies.
The multivariable models used in this study are an exercise in decision making, and appropriate
interpretation of the results is required because the events rate was very low. The inclusion of the
modelling work within the thesis, however, is beneficial for future research.
Another strength of this study is that parents were included in VAP education. In doing so, this study
involved parents in one of the key VAP preventative strategies: hand hygiene. The decision to involve
parents in this strategy arose from discussions with respective clinicians, educators and researchers.
This particular partnership between nurses and parents is also recommended by the World Health
Organisation (World Health Organization (WHO), 2009b). No survey or hand hygiene auditing
among parents in PICU had been undertaken before this study. More studies are needed to set the
hand hygiene benchmark for parents and visitors in PICUs.
This study successfully used appropriate methodologies to comprehensively examine VAP/VAE in
PICUs and define measures to overcome the existing challenges in preventing VAP/VAE in children.
164
The retrospective study enabled evaluation of the magnitude of VAP in a PICU, and at the same time
tested the applicability of the new VAE surveillance tool. Although the retrospective study in Phase
1 did not directly inform Phase 2; VAP education, compliance auditing with feedback and surveys,
the relevancy of VAP education involving PICU staff and parents, compliance, and feedback were
found to be crucial elements in minimising the impact of VAP in PICUs. The study concluded by
evaluating possible improvements for implementation after reinforcement of VAP education, and
compliance auditing with feedback in a six-month prospective study where patients were followed to
observe their progress.
General implications and future directions
The present study contributes to VAP epidemiological data in children and also adds to the growing
body of clinical knowledge on VAE surveillance in critically ill children receiving mechanical
ventilation in PICUs. In the Australian context, this is the first time that the CDC’s VAE surveillance
tool for adults has been applied to paediatric patients. This study has been cautious in applying VAE
tools to historical data, as the data gathered may not be wholly accurate due to the method in which
it was collected (via retrospective studies) (Phongjitsiri et al., 2015; Stein et al., 2014).
Retrieving data from medical records can be challenging, especially when data is not electronically
recorded, and automatic data retrieval is not available (Cocoros, Priebe, Logan, et al., 2017). It is
expected that the application of VAE surveillance tools will be mandated in children, although the
present evidence on paediatric versions of VAE surveillance version is still insufficient (Mohd Ali et
al., 2019). The ready availability of the clinical data in the electronic medical records made
surveillance of VAE in children possible and may have been able to minimise inter-observer bias as
well as increase the accuracy of data collected (as compared to what might be gathered via the manual
method) (Hebert et al., 2017; Stevens et al., 2014). The electronic medical record Metavision
(iMDsoft®), used in this study is not able to perform an automatic retrieval of the clinical data for
VAE surveillance purposes, but this presents an opportunity for future collaboration between
information technology personnel and clinicians in designing automatic data retrieval, which could
reduce the amount of time required to collate data (Hebert et al., 2017; Stevens et al., 2014). Since
the VAE surveillance criteria consists of objective data which is readily available within electronic
medical records, the reliability of the data obtained has increased (Klompas, 2013a; Klompas et al.,
2011; Magill et al., 2013; Stevens et al., 2014). Further studies may find it useful to develop an
algorithm to capture elements of VAE tiers in children.
165
The tools of VAP prevention strategies compliance (VAP bundle) and individual VAP prevention
strategies need to be clearly defined to enable inter-institutional comparison and benchmarking.
Similarly, the benchmark for parents’ and visitors’ hand hygiene compliance needs to be clearly
defined. This classification of compliance is vital, yet noticeably missing in current literature. This
study proposes the need for hand hygiene compliance monitoring among them as similarly
highlighted in a recent study by Giannini et al. (2016). Thus, future research may focus on this area.
VAP education in this study was delivered through online learning (TEACHQ platform), an internal
learning platform available within the hospital. The study used this approach to maximise the
engagement of PICU staff with VAP education materials (Blot, Koulenti, & Labeau, 2017; Cirulis et
al., 2016; Labeau et al., 2016), but assessing the effectiveness of this method was not a focus of this
study, so it is recommended that future research may assess the usefulness of this method.
Conclusion
The findings of this study demonstrated a reduction of VAP/VAE in a six-month prospective study
defined by two surveillance tools. Although generalisation is limited, this study demonstrated that a
decrease of VAP/VAE incidence rates can occur in a PICU after updated VAP education is
undertaken through educating PICU staff and parents, coupled with compliance auditing and
feedback in the PICU. This study demonstrated that these changes in the process of care for the
patients in PICU improved patient outcomes in terms of reduced rates of VAP and VAE, although
this reduction may also be influenced by the case-mix. This study contributes to and extends the work
of previous literature, especially the understanding of VAP preventative strategies tailored to
paediatrics. The new VAE surveillance tools applied to 2015 patient cohorts revealed a similar
incidence rate, although very small differences were reported when the incidence calculation was
based on the patients no longer at risk. Although poor agreement between PNU/VAP and the new
VAE surveillance tool and low sensitivity were found, the high specificity suggests the merit of the
new VAE tools to identify complications other than VAP.
While a multivariate analysis in this study found no significant association between the tested
potential risk factors and VAP preventative strategies in VAP models, the VAE model results showed
that frequent oral hygiene performance and the presence of GI prophylaxis was associated with VAE
development. This offers insight into current practises in PICUs, and these need further investigation.
The significant associations at univariate analyses and/or insignificant results for VAP show that VAP
is a serious complication in PICUs and requires re-evaluation in terms of individual VAP preventative
strategies. The present study utilised updated VAP education and compliance auditing with feedback
166
involving both PICU staff and parents. The surveys on ‘Speaking up for hand hygiene’ among parents
and nurses suggested that agreement on this initiative would increase good hand hygiene practise.
This study also highlighted perceptions of staff and parents about reminding each other to perform
hand hygiene. Further research exploring this relationship could also promote partnerships to ensure
optimal hand hygiene practise and prevent infection in PICUs. Significant compliance improvement
for VAP preventative strategies was achieved between retrospective and prospective studies, and this
suggests the usefulness of updated education and compliance auditing with feedback, as these were
found to directly aid the reduction of VAP/VAE incidences in PICU.
167
List of References
Aelami, M. H., Lotfi, M., & Zingg, W. (2014). Ventilator-associated pneumonia in neonates, infants
and children. Antimicrobial Resistance and Infection Control, 3(1). doi:10.1186/2047-2994-
3-30
Agarwal, S., Burke, P., Bird, D., O’Donnell, C., Zambuto, A., Burke, R., .. Korn, C. (2010).
Adherence to ventilator-associated pneumonia bundle and incidence of ventilator-associated
pneumonia in the surgical intensive care unit. Archives of Surgery, 145(5), 465-470.
doi:10.1001/archsurg.2010.69
Al-Mousa, H. H., Omar, A. A., Rosenthal, V. D., Salama, M. F., Aly, N. Y., El-Dossoky Noweir, M.,
Varghese, S. T. (2016). Device-associated infection rates, bacterial resistance, length of stay,
and mortality in Kuwait: International Nosocomial Infection Consortium findings. American
Journal of Infection Control, 44(4), 444-449. doi:10.1016/j.ajic.2015.10.031
Al-Thaqafy, M. S., El-Saed, A., Arabi, Y. M., & Balkhy, H. H. (2014). Association of compliance of
ventilator bundle with incidence of ventilator-associated pneumonia and ventilator utilization
among critical patients over 4 years. Annal Thoracic Medicine, 9(4), 221-226.
doi:10.4103/1817-1737.140132
Albert, D. B., Zurakowski, J. D., Bechard, P. L., Priebe, P. G., Duggan, K. C., Heyland, M. D., &
Mehta, M. N. (2016). Enteral nutrition and acid-suppressive therapy in the PICU: Impact on
the risk of ventilator-associated pneumonia. Pediatric Critical Care Medicine, 17(10), 924-
929. doi:10.1097/PCC.0000000000000915
Allegranzi, B., Gayet-Ageron, A., Damani, N., Bengaly, L., McLaws, M.-L., Moro, M.-L., . . . Pittet,
D. (2013). Global implementation of WHO's multimodal strategy for improvement of hand
hygiene: A quasi-experimental study. The Lancet Infectious Diseases, 13(10), 843-851.
doi:10.1016/S1473-3099(13)70163-4
Allen Furr, L., Binkley, C. J., McCurren, C., & Carrico, R. (2004). Factors affecting quality of oral
care in intensive care units. Journal of Advanced Nursing, 48(5), 454-462.
doi:10.1111/j.1365-2648.2004.03228.x
Allison, J. J., Wall, T. C., Spettell, C. M., Calhoun, J., Fargason Jr, C. A., Kobylinski, R. W., . . .
Kiefe, C. (2000). The art and science of chart review. The Joint Commission Journal on
Quality Improvement, 26(3), 115-136. Retrieved from https://www.journals.elsevier.com/the-
joint-commission-journal-on-quality-and-patient-safety.
Almuneef, M., Memish, Z. A., Balkhy, H. H., Alalem, H., & Abutaleb, A. (2004). Ventilator-
associated pneumonia in a pediatric intensive care unit in Saudi Arabia: A 30-month
prospective surveillance. Infection Control & Hospital Epidemiology, 25.
doi:10.1086/502472
Altman, D. G., & Bland, J. M. (1994). Statistics Notes: Diagnostic tests 1: Sensitivity and specificity.
BMJ, 308(6943), 1552. doi:10.1136/bmj.308.6943.1552
American Thoracic, S., & Infectious Diseases Society, A. (2005). Guidelines for the management of
adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia.
American Journal of Respiratory and Critical Care Medicine, 171(4), 388-416.
doi:10.1164/rccm.200405-644ST
Ames, N. J. (2011). Evidence to support tooth brushing in critically ill patients. American Journal of
Critical Care: An Official Publication, American Association of Critical-Care Nurses, 20(3),
242-250. doi:10.4037/ajcc2011120
Amin, A. N., & Deruelle, D. (2015). Healthcare-associated infections, infection control and the
potential of new antibiotics in development in the USA. Future Microbiology, 10(6), 1049-
1062. doi:10.2217/fmb.15.33
Anthony, M., Bedford-Russell, A., Cooper, T., Fry, C., Heath, P. T., Kennea, N., . . . Wilson, P.
(2013). Managing and preventing outbreaks of Gram-negative infections in UK neonatal
168
units. Archives of Disease Childhood Fetal Neonatal Edition, 98(6), F549-553.
doi:10.1136/archdischild-2012-303540
Arabi, Y., Al-Shirawi, N., Memish, Z., & Anzueto, A. (2008). Ventilator-associated pneumonia in
adults in developing countries: A systematic review. International Journal of Infectious
Diseases, 12(5), 505-512. doi:10.1016/j.ijid.2008.02.010
Araujo, T. E., Vieira, S. M. G., & Carvalho, P. R. A. (2010). Stress ulcer prophylaxis in pediatric
intensive care units. Jornal de Pediatria, 86(6), 525. doi:doi:10.2223/JPED.2046
Arca, M. J., Uhing, M., & Wakeham, M. (2015). Current concepts in acute respiratory support for
neonates and children. Seminars in Pediatric Surgery, 24(1), 2-7.
doi:10.1053/j.sempedsurg.2014.11.001
Ashton, C. M., Del Junco, D. J., Souchek, J., Wray, N. P., & Mansyur, C. L. (1997). The Association
between the Quality of Inpatient Care and Early Readmission: A Meta-Analysis of the
Evidence. Medical Care, 35(10), 1044-1059. doi:10.1097/00005650-199710000-00006
Australian and New Zealand Intensive Care Society. (2016). CORE Annual Reports. Retrieved from
https://www.anzics.com.au/wp-content/uploads/2018/08/ANZPICR-Annual-Report-
2015.pdf
Australian and New Zealand Intensive Care Society. (2017). CORE Annual Reports. Retrieved from
https://www.anzics.com.au/wp-content/uploads/2018/08/ANZPICR-Annual-Report-
2016.pdf
Australian Bureau of Statistics. (2009). Health Literacy. Retrieved from
http://www.abs.gov.au/AUSSTATS/[email protected]/Lookup/4102.0Main+Features20June+2009
Australian Commission on safety and Quality in Health Care. National Hand Hygiene. (2013).
National Hand Hygiene Initiative. Retrieved from https://www.safetyandquality.gov.au/
Auxiliadora-Martins, M., Menegueti, M. G., Nicolini, E. A., Alkmim-Teixeira, G. C., Bellissimo-
Rodrigues, F., Martins-Filho, O. A., & Basile-Filho, A. (2012). Effect of heat and moisture
exchangers on the prevention of ventilator-associated pneumonia in critically ill patients.
Brazilian Journal of Medical and Biological Research, 45(12), 1295-1300. doi:
10.1590/S0100-879X2012007500161
Awasthi, S., Tahazzul, M., Ambast, A., Govil, Y. C., & Jain, A. (2013). Longer duration of
mechanical ventilation was found to be associated with ventilator-associated pneumonia in
children aged 1 month to 12 years in India. Journal of Clinical Epidemiology, 66(1), 62-66.
doi:10.1016/j.jclinepi.2012.06.006
Azab, S. F. A., Sherbiny, H. S., Saleh, S. H., Elsaeed, W. F., Elshafiey, M. M., Siam, A. G., . . .
Gheith, T. (2015). Reducing ventilator-associated pneumonia in neonatal intensive care unit
using "VAP prevention Bundle": A cohort study. BMC Infectious Diseases, 15, 314-314.
doi:10.1186/s12879-015-1062-1
Babcock, H. M., Zack, J. E., Garrison, T., Trovillion, E., Jones, M., Fraser, V. J., & Kollef, M. H.
(2004). An educational intervention to reduce ventilator-associated pneumonia in an
integrated health system: A comparison of effects. Chest, 125(6), 2224-2231.
doi:10.1378/chest.125.6.2224
Balasubramanian, P., & Tullu, M. (2014). Study of ventilator-associated pneumonia in a pediatric
intensive care unit. The Indian Journal of Pediatrics, 81(11), 1182-1186. doi:10.1007/s12098-
014-1444-1
Barr, J., Hecht, M., Flavin, K. E., Khorana, A., & Gould, M. K. (2004). Outcomes in critically ill
patients before and after the implementation of an evidence-based nutritional management
protocol. Chest, 125(4), 1446-1457. doi:10.1378/chest.125.4.1446
Bassant Salah, M., Sally, K., & Seham Awad El, S. (2017). Mechanical ventilation practise in
Egyptian pediatric intensive care units. Electronic physician, 9(5), 4370-4377.
doi:10.19082/4370
Beardsley, A. L., Nitu, M. E., Cox, E. G., & Benneyworth, B. D. (2016). An evaluation of various
ventilator-associated infection criteria in a PICU. Pediatric Critical Care Medicine, 17(1), 73-
80. doi:10.1097/PCC.0000000000000569
169
Becerra, M. R., Tantalean, J. A., Suarez, V. J., Alvarado, M. C., Candela, J. L., & Urcia, F. C. (2010).
Epidemiologic surveillance of nosocomial infections in a pediatric intensive care unit of a
developing country. BMC Pediatrics, 10. doi:10.1186/1471-2431-10-66
Belela-Anacleto, A. S. C., Kusahara, D. M., Peterlini, M. A. S., & Pedreira, M. L. G. (2019). Hand
hygiene compliance and behavioural determinants in a paediatric intensive care unit: An
observational study. Australian Critical Care, 32(1), 21-27. doi:10.1016/j.aucc.2018.02.010
Bellissimo-Rodrigues, F., Bellissimo-Rodrigues, W. T., & Menegueti, M. G. (2014). Selfishness
among healthcare workers and nosocomial infections: A causal relationship? Revista da
Sociedade Brasileira de Medicina Tropical, 47(4), 407. doi:10.1590/0037-8682-0191-201
Bellissimo-Rodrigues, F., Pires, D., Zingg, W., & Pittet, D. (2016). Role of parents in the promotion
of hand hygiene in the paediatric setting: A systematic literature review. Journal of Hospital
Infection, 93(2), 159-163. doi:10.1016/j.jhin.2016.02.001
Benet, T., Allaouchiche, B., Argaud, L., & Vanhems, P. (2012). Impact of surveillance of hospital-
acquired infections on the incidence of ventilator-associated pneumonia in intensive care
units: A quasi-experimental study. Critical Care, 16(4), R161. doi:10.1186/cc11484
Berry, A. M., Davidson, P. M., Masters, J., & Rolls, K. (2007). Systematic literature review of oral
hygiene practises for intensive care patients receiving mechanical ventilation. American
Journal of Critical Care, 16(6), 552-562; quiz 563. Retrieved from
https://www.ncbi.nlm.nih.gov/pubmed/17962500.
Berry, A. M., Davidson, P. M., Masters, J., Rolls, K., & Ollerton, R. (2011). Effects of three
approaches to standardized oral hygiene to reduce bacterial colonization and ventilator
associated pneumonia in mechanically ventilated patients: A randomised control trial.
International Journal of Nursing Studies, 48(6), 681-688. doi:10.1016/j.ijnurstu.2010.11.004
Berry, A. M., Davidson, P. M., Nicholson, L., Pasqualotto, C., & Rolls, K. (2011). Consensus based
clinical guideline for oral hygiene in the critically ill. Intensive and Critical Care Nursing,
27(4), 180-185. doi:10.1016/j.iccn.2011.04.005
Bhardwaj, N. (2013). Pediatric cuffed endotracheal tubes. Journal of Anaesthesiology Clinical
Pharmacology, 29(1), 13-18. doi:10.4103/0970-9185.105786
Bigham, M. T., Amato, R., Bondurrant, P., Fridriksson, J., Krawczeski, C. D., Raake, J., . . . Brilli,
R. J. (2009). Ventilator-associated pneumonia in the pediatric intensive care unit:
Characterizing the problem and implementing a sustainable solution. Journal of Pediatrics,
154(4), 582-587 e582. doi:10.1016/j.jpeds.2008.10.019
Blot, S., Koulenti, D., & Labeau, S. (2017). Optimizing educational initiatives to prevent ventilator-
associated complications. AJIC: American Journal of Infection Control, 45(1), 102-103.
doi:10.1016/j.ajic.2016.05.035
Borgert, M. J., Goossens, A., & Dongelmans, D. A. (2015). What are effective strategies for the
implementation of care bundles on ICUs: A systematic review. Implementation Science,
10(1). doi:10.1186/s13012-015-0306-1
Boscart, V. M., Fernie, G. R., Lee, J. H., & Jaglal, S. B. (2012). Using psychological theory to inform
methods to optimize the implementation of a hand hygiene intervention. Implement Science,
7(1), 77. doi:10.1186/1748-5908-7-77
Boscart, V. M., McGilton, K. S., Levchenko, A., Hufton, G., Holliday, P., & Fernie, G. R. (2008).
Acceptability of a wearable hand hygiene device with monitoring capabilities. Journal of
Hospital Infection, 70(3), 216-222. doi:10.1016/j.jhin.2008.07.008
Bouadma, L., Mourvillier, B., Deiler, V., Derennes, N., Le Corre, B., Lolom, I., . . . Lucet, J.-C.
(2010). Changes in knowledge, beliefs, and perceptions throughout a multifaceted behavioral
program aimed at preventing ventilator-associated pneumonia. Intensive Care Medicine,
36(8), 1341-1347. doi:10.1007/s00134-010-1890-9
Bouadma, L., Mourvillier, B., Deiler, V., Le Corre, B., Lolom, I., Regnier, B., . . . Lucet, J. C. (2010).
A multifaceted program to prevent ventilator-associated pneumonia: impact on compliance
with preventive measures. Critical Care Medicine, 38(3), 789-796.
doi:10.1097/CCM.0b013e3181ce21af
170
Bouadma, L., Sonneville, R., Garrouste-Orgeas, M., Darmon, M., Souweine, B., Voiriot, G., . . .
Group, O. S. (2015). Ventilator-Associated Events: Prevalence, outcome, and relationship
with ventilator-associated pneumonia. Critical Care Medicine, 43(9), 1798-1806.
doi:10.1097/CCM.0000000000001091
Boyer, A. F., Schoenberg, N., Babcock, H., McMullen, K. M., Micek, S. T., & Kollef, M. H. (2015).
A prospective evaluation of ventilator-associated conditions and infection-related ventilator-
associated conditions. Chest, 147(1), 68-81. doi:10.1378/chest.14-0544
Brierley, J., Highe, L., Hines, S., & Dixon, G. (2012). Reducing VAP by instituting a care bundle
using improvement methodology in a UK paediatric intensive care unit. European Journal of
Pediatrics, 171(2), 323-330. doi:10.1007/s00431-011-1538-y
Brilli, R. J., Sparling, K. W., Lake, M. R., Butcher, J., Myers, S. S., Clark, M. D., . . . Stutler, M. E.
(2008). The business case for preventing ventilator-associated pneumonia in pediatric
intensive care unit patients. The Joint Commission Journal on Quality and Patient Safety,
34(11), 629-638. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/19025083.
Bsharat, S., & Drach-Zahavy, A. (2017). Nurses' response to parents' 'speaking-up' efforts to ensure
their hospitalized child's safety: An attribution theory perspective. Journal of Advanced
Nursing, 73(9), 2118-2128. doi:10.1111/jan.13282
Buser, G. L., Fisher, B. T., Shea, J. A., & Coffin, S. E. (2013). Parent willingness to remind health
care workers to perform hand hygiene. American Journal of Infection Control, 41(6), 492-
496. doi:10.1016/j.ajic.2012.08.006
Casado, R. J. A., de Mello, M. J. G., de Aragão, R. C. F., de Albuquerque, M. d. F. P. M., & Correia,
J. B. (2011). Incidence and risk factors for health care-associated pneumonia in a pediatric
intensive care unit. Critical Care Medicine, 39(8), 1968-1973.
doi:10.1097/CCM.0b013e31821b840d
Cason, C. L., Tyner, T., Saunders, S., Broome, L., Centers for Disease, C., & Prevention. (2007).
Nurses' implementation of guidelines for ventilator-associated pneumonia from the Centers
for Disease Control and Prevention. American Journal of Critical Care, 16(1), 28-36;
discussion 37; quiz 38. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/17192524.
Centers for Disease Control and Prevention (CDC). (2016a, September 13). Epi InfoTM. Retrieved
from https://www.cdc.gov/epiinfo/index.html
Centers for Disease Control and Prevention (CDC). (2016b, April 24). NHSN Ventilator_Associated
Event (VAE) Calculator Ver. 4.0. Retrieved from
https://nhsn.cdc.gov/nhsntraining/vae4/vaecalc_v4.html
Centers for Disease Control and Prevention (CDC) & National Healthcare Safety Network (NHSN).
(2015a). Pneumonia (Ventilator –associated [VAP] and non- ventilator associated
pneumonia [PNEU] event Protocol. Retrieved from
https://www.cdc.gov/nhsn/pdfs/pscmanual/6pscvapcurrent.pdf
Centers for Disease Control and Prevention (CDC) & National Healthcare Safety Network (NHSN).
(2015b). Surveillance for Ventilator-associated Event (VAE) Protocol. Retrieved from
https://www.cdc.gov/nhsn/pdfs/pscmanual/10-vae_final.pdf
Chandonnet, J. C., Boutwell, M. K., Spigel, M. N., Carter, M. J., Degrazia, M. M., Ozonoff, M. A.,
& Flaherty, M. K. (2017). Itʼs in Your Hands: An educational initiative to improve
parent/family hand hygiene compliance. Dimensions of Critical Care Nursing, 36(6), 327-
333. doi:10.1097/DCC.0000000000000268
Chang, E., Easterbrook, S., Hancock, K., Johnson, A., & Davidson, P. (2010). Evaluation of an
information booklet for caregivers of people with dementia: An Australian perspective.
Nursing & Health Sciences, 12(1), 45-51. doi:10.1111/j.1442-2018.2009.00486.x
Chang, I., & O'Leary, J. (2016). Ventilator Associated Pneumonia Prevention: Nursing Interventions.
Unpublished manuscript. Lady Cilento Children’s Hospital, Brisbane, Australia.
Chang, I., & Schibler, A. (2016). Ventilator Associated Pneumonia in Children. Paediatric
Respiratory Review, 20, 10-16. doi:10.1016/j.prrv.2015.09.005
171
Charles, M. P., Kali, A., Easow, J. M., Joseph, N. M., Ravishankar, M., Srinivasan, S., . .Umadevi,
S. (2014). Ventilator-associated pneumonia. Australasian Medical Journal, 7(8), 334-344.
doi:10.4066/AMJ.2014.2105
Chastre, J., & Fagon, J. Y. (2002). Ventilator-associated pneumonia. American Journal of
Respiratory and Critical Care Medicine, 165(7), 867-903.
doi:10.1164/ajrccm.165.7.2105078
Chastre, J., Wolff, M., Fagon, J.-Y., Chevret, S., Thomas, F., Wermert, D., . . . for the Pneuma Trial,
G. (2003). Comparison of 8 vs 15 days of antibiotic therapy for ventilator-associated
pneumonia in adults: A randomized trial. JAMA, 290(19), 2588-2598.
doi:10.1001/jama.290.19.2588
Chen, L. F., Carriker, C., Staheli, R., Isaacs, P., Elliott, B., Miller, B. A., . . Sexton, D. J. (2013).
Observing and Improving Hand Hygiene Compliance: Implementation and refinement of an
Electronic-Assisted Direct-Observer Hand Hygiene Audit Program. Infection Control and
Hospital Epidemiology, 34(2), 207-210. doi:10.1086/669084
Chen, Y. C., & Chiang, L. C. (2007). Effectiveness of hand-washing teaching programs for families
of children in paediatric intensive care units. Journal of Clinical Nursing, 16(6), 1173-1179.
doi:10.1111/j.1365-2702.2007.01665.x
Chhapola, V., & Brar, R. (2015). Impact of an educational intervention on hand hygiene compliance
and infection rate in a developing country neonatal intensive care unit. International Journal
of Nursing Practise, 21(5), 486-492. doi:10.1111/ijn.12283
Children's Health Queensland. (2016). Welcome to LCCH book. Retrieved from
http://www.childrens.health.qld.gov.au/wp-content/uploads/2014/11/welcome-lcch-
book.pdf
Children’s Health Queensland Hospital and Health Service. (2016). Patient Safety. Retrieved from
https://www.childrens.health.qld.gov.au/chq/about-us/patient-safety/
Children’s Health Queensland Hospital and Health Service. (2017). Patient Safety. Retrieved from
https://www.childrens.health.qld.gov.au/chq/about-us/patient-safety/
Choules, A. P. (2007). The use of elearning in medical education: A review of the current situation.
Postgraduate Medical Journal, 83(978), 212. doi:10.1136/pgmj.2006.054189
Chu, S.-M., Yang, M.-C., Hsiao, H.-F., Hsu, J.-F., Lien, R., Chiang, M.-C., . . . Tsai, M.-H. (2015).
One-week versus 2-day ventilator circuit change in neonates with prolonged ventilation: Cost-
effectiveness and impact on ventilator-associated pneumonia. Infection Control and Hospital
Epidemiology, 36(3), 287-293. doi:10.1017/ice.2014.48
Ciofi degli Atti, M. L., Tozzi, A. E., Ciliento, G., Pomponi, M., Rinaldi, S., & Raponi, M. (2011).
Healthcare workers' and parents' perceptions of measures for improving adherence to hand-
hygiene.(Research article). BMC Public Health, 11, 466. doi: 10.1186/1471-2458-11-466
Cirulis, M. M., Hamele, M. T., Stockmann, C. R., Bennett, T. D., & Bratton, S. L. (2016). Comparison
of the new adult ventilator-associated event criteria to the Centers for Disease Control and
Prevention pediatric ventilator-associated pneumonia definition (PNU2) in a population of
pediatric traumatic brain injury patients. Pediatric Critical Care Medicine, 17(2), 157-164.
doi:10.1097/PCC.0000000000000590
Clark, M. (2008). Retrospective versus prospective cohort study designs for evaluating treatment of
pressure ulcers: A comparison of 2 studies. Journal of Wound, Ostomy and Continence
Nursing, 35(4), 391-394. doi:10.1097/01.WON.0000326657.57939.c9
Cocoros, N. M., Kleinman, K., Priebe, G. P., Gray, J. E., Logan, L. K., Larsen, G., . . . Pediatric
Ventilator-Associated Conditions Study, T. (2016). Ventilator-associated events in neonates
and children: A new paradigm. Critical Care Medicine, 44(1), 14-22.
doi:10.1097/CCM.0000000000001372
Cocoros, N. M., Priebe, G., Gray, J. E., Toltzis, P., Larsen, G. Y., Logan, L. K., . . . Lee, G. M. (2017).
Factors associated with pediatric ventilator-associated conditions in six U.S. hospitals: A
nested case-control study. Pediatric Critical Care Medicine, 18(11), e536-e545.
doi:10.1097/PCC.0000000000001328
172
Cocoros, N. M., Priebe, G. P., Logan, L. K., Coffin, S., Larsen, G., Toltzis, P., . . . Lee, G. M. (2017).
A pediatric approach to ventilator-associated events surveillance. Infection Control &
Hospital Epidemiology, 38(3), 327-333. doi:10.1017/ice.2016.277
Coffin, S. E., Klompas, M., Classen, D., Arias, K. M., Podgorny, K. R., Anderson, D. J., ... Yokoe,
D. S. (2008). Strategies to prevent ventilator-associated pneumonia in acute care hospitals.
Infection Control and Hospital Epidemiology, 29(S1), S31-S40. doi:10.1086/591062
Cook, D. J., & Kollef, M. H. (1998). Risk factors for ICU acquired pneumonia. JAMA, 279(20), 1605-
1606. doi:10.1001/jama.279.20.1605
Cooper, V. B., & Haut, C. (2013). Preventing ventilator-associated pneumonia in children: an
evidence-based protocol. Critical Care Nurse, 33(3), 21-29; quiz 30. doi:10.4037/ccn2013204
Copnell, B., Tingay, D. G., Kiraly, N. J., Sourial, M., Gordon, M. J., Mills, J. F., Dargaville, P. A.
(2007). A comparison of the effectiveness of open and closed endotracheal suction. Intensive
Care Medicine, 33(9), 1655-1662. doi: 10.1007/s00134-007-0635-x
Corlett, J., & Twycross, A. (2006). Negotiation of parental roles within Family‐Centred Care: a
review of the research. Journal of Clinical Nursing, 15(10), 1308-1316. doi:10.1111/j.1365-
2702.2006.01407.x
Coyne, I., Murphy, M., Costello, T., O’Neill, C., & Donnellan, C. (2013). A survey of nurses’
practises and perceptions of Family-Centered Care in Ireland. Journal of Family Nursing,
19(4), 469-488. doi:10.1177/1074840713508224
Craven, D. E. (2006). Preventing ventilator-associated pneumonia in adults: Sowing seeds of change.
Chest, 130(1), 251-260. doi:10.1378/chest.130.1.251
Craven, D. E., Chroneou, A., Zias, N., & Hjalmarson, K. I. (2009). Ventilator-associated
tracheobronchitis: The impact of targeted antibiotic therapy on patient outcomes: Chest,
135(2), 521-528. doi:10.1378/chest.08-1617
Curley, M. A., Schwalenstocker, E., Deshpande, J. K., Ganser, C. C., Bertoch, D., Brandon, J., &
Kurtin, P. (2006). Tailoring the Institute for Health Care Improvement 100,000 Lives
Campaign to pediatric settings: The example of ventilator-associated pneumonia. Pediatric
Clinics of North America, 53(6), 1231-1251. doi:10.1016/j.pcl.2006.09.001
Daniels, J. P., Hunc, K., Cochrane, D. D., Carr, R., Shaw, N. T., Taylor, A., . . . Ansermino, J. M.
(2012). Identification by families of pediatric adverse events and near misses overlooked by
health care providers. CMAJ: Canadian Medical Association Journal, 184(1), 29-34.
doi:10.1503/cmaj.110393
Dave, M. H., Frotzler, A., Madjdpour, C., Koepfer, N., & Weiss, M. (2011). Massive aspiration past
the tracheal tube cuff caused by closed tracheal suction system. Journal of Intensive Care
Medicine, 26(5), 326-329. doi:10.1177/0885066610392516
Davies, K., Monterosso, L., & Leslie, G. (2011). Determining standard criteria for endotracheal
suctioning in the paediatric intensive care patient: An exploratory study. Intensive & Critical
Care Nursing, 27(2), 85-93. doi:10.1016/j.iccn.2011.01.002
Davis, R., Parand, A., Pinto, A., & Buetow, S. (2015). Systematic review of the effectiveness of
strategies to encourage patients to remind healthcare professionals about their hand hygiene.
Journal of Hospital Infection, 89(3), 141-162. doi:10.1016/j.jhin.2014.11.010
De Cristofano, A., Peuchot, V., Canepari, A., Franco, V., Perez, A., & Eulmesekian, P. (2016).
Implementation of a ventilator-associated pneumonia prevention bundle in a single PICU.
Pediatric Critical Care Medicine, 17(5), 451-456. doi:10.1097/pcc.0000000000000714
Dhar, S., Tansek, R., Toftey, Elizabeth A., Dziekan, Beth A., Chevalier, Thomas C., Bohlinger,
Connie G., . . . Kaye, Keith S. (2010). Observer bias in hand hygiene compliance reporting.
Infection Control and Hospital Epidemiology, 31(8), 869-870. doi:10.1086/655441
Dixit, D., Hagtvedt, R., Reay, T., Ballermann, M., & Forgie, S. (2012). Attitudes and beliefs about
hand hygiene among paediatric residents: A qualitative study. BMJ Open, 2(6).
doi:10.1136/bmjopen-2012-002188
Doig, G., Heighes, P., Simpson, F., Sweetman, E., & Davies, A. (2009). Early enteral nutrition,
provided within 24 h of injury or intensive care unit admission, significantly reduces mortality
173
in critically ill patients: A meta-analysis of randomised controlled trials. Intensive Care
Medicine, 35(12), 2018-2027. doi:10.1007/s00134-009-1664-4
Drakulovic, M. B., Torres, A., Bauer, T. T., Nicolas, J. M., Nogué, S., & Ferrer, M. (1999). Supine
body position as a risk factor for nosocomial pneumonia in mechanically ventilated patients:A
randomised trial. The Lancet, 354(9193), 1851-1858. doi:10.1016/S0140-6736(98)12251-1
Dudeck, M. A., Horan, T. C., Peterson, K. D., Allen-Bridson, K., Morrell, G. C., Pollock, D. A., &
Edwards, J. R. (2011). National Healthcare Safety Network (NHSN) report, data summary for
2009, device-associated module. AJIC: American Journal of Infection Control, 39(5), 349-
367. doi:10.1016/j.ajic.2011.04.011
Duffett, M., Choong, K., Foster, J., Gilfoyle, E., Lacroix, J., Pai, N., . . . Canadian Critical Care Trials,
G. (2017). Pediatric intensive care stress ulcer prevention (PIC-UP): A protocol for a pilot
randomized trial. Pilot and Feasibility Studies, 3(1), 26. doi:10.1186/s40814-017-0142-y
Duffett, M., Choong, K., Hartling, L., Menon, K., Thabane, L., & Cook, D. J. (2013). Randomized
controlled trials in pediatric critical care: a scoping review. Critical Care, 17(5), R256.
doi:10.1186/cc13083
Dullenkopf, A., Bernet‐Buettiker, V., Maino, P., & Weiss, M. (2006). Performance of a novel
pressure release valve for cuff pressure control in pediatric tracheal tubes. Pediatric
Anesthesia, 16(1), 19-24. doi:10.1111/j.1460-9592.2005.01686.x
Eilers, J., Berger, A. M., & Petersen, M. C. (1988). Development, testing, and application of the oral
assessment guide. Oncology Nursing Forum, 15(3), 325-330. Retrieved from
https://www.ncbi.nlm.nih.gov/pubmed/3287344.
Elward, A. M., Warren, D. K., & Fraser, V. J. (2002). Ventilator-Associated Pneumonia in Pediatric
Intensive Care Unit Patients: Risk factors and outcomes. Pediatrics, 109(5), 758-764.
doi:10.1542/peds.109.5.758
Eom, J. S., Lee, M. S., Chun, H. K., Choi, H. J., Jung, S. Y., Kim, Y. S., . . . Lee, J. S. (2014). The
impact of a ventilator bundle on preventing ventilator-associated pneumonia: A multicenter
study. American Journal of Infection Control, 42(1), 34-37. doi:10.1016/j.ajic.2013.06.023
Erasmus, V., Daha, Thea J., Brug, H., Richardus, Jan H., Behrendt, Myra D., Vos, Margreet C., &
van Beeck, Ed F. (2010). Systematic review of studies on compliance with hand hygiene
guidelines in hospital care. Infection Control and Hospital Epidemiology, 31(3), 283-294.
doi:10.1086/650451
Eschertzhuber, S., Salgo, B., Schmitz, A., Roth, W., Frotzler, A., Keller, C. H., . . . Weiss, M. (2010).
Cuffed endotracheal tubes in children reduce sevoflurane and medical gas consumption and
related costs. Acta Anaesthesiologica Scandinavica, 54(7), 855-858. doi:10.1111/j.1399-
6576.2010.02261.x
Evans, J., Syddall, S., Butt, W., & Kinney, S. (2014). Comparison of open and closed suction on
safety, efficacy and nursing time in a paediatric intensive care unit. Australia Critical Care,
27(2), 70-74; quiz 75-76. doi:10.1016/j.aucc.2014.01.003
Fitzpatrick, M., Everett‐Thomas, R., Nevo, I., Shekhter, I., Rosen, L. F., Scheinman, S. R., . . .
Birnbach, D. J. (2011). A novel educational programme to improve knowledge regarding
health care‐associated infection and hand hygiene. International Journal of Nursing
Practise, 17(3), 269-274. doi:10.1111/j.1440-172X.2011.01934.x
Flodgren, G., Conterno, L. O., Mayhew, A., Omar, O., Pereira, C. R., & Shepperd, S. (2013).
Interventions to improve professional adherence to guidelines for prevention of device-related
infections. Cochrane Database of Systematic Reviews, 3(3), CD006559.
doi:10.1002/14651858.CD006559.pub2
Foglia, E., Meier, M. D., & Elward, A. (2007). Ventilator-associated pneumonia in neonatal and
pediatric intensive care unit patients. Clinical Microbiology Reviews, 20(3), 409-425, table of
contents. doi:10.1128/CMR.00041-06
Ford-Jones, L. E., Mindorff, M. C., Langley, M. J., Allen, L. U., Nàvàs, L. L., Patrick, L. M., Gold,
L. R. (1989). Epidemiologic study of 4684 hospital-acquired infections in pediatric patients.
174
The Pediatric Infectious Disease Journal, 8(10), 668-675. doi:10.1097/00006454-
198910000-00002
Fourrier, F., Dubois, D., Pronnier, P., Herbecq, P., Leroy, O., Desmettre, T., Group, P. S. (2005).
Effect of gingival and dental plaque antiseptic decontamination on nosocomial infections
acquired in the intensive care unit: A double-blind placebo-controlled multicenter study.
Critical Care Medicine, 33(8), 1728-1735. doi: 10.1097/01.CCM.0000171537.03493.B0
Friedman, N. D., Russo, P. L., & Richards, M. J. (2005). Surveillance for ventilator-associated
pneumonia: the challenges and pitfalls. Australian Infection Control, 10(4), 122-125.
doi:10.1071/hi05122
Gandra, S., & Ellison, R. T., 3rd. (2014). Modern trends in infection control practises in intensive
care units. Journal of Intensive Care Medicine, 29(6), 311-326.
doi:10.1177/0885066613485215
Garner, J. S., Jarvis, W. R., Emori, T. G., Horan, T. C., & Hughes, J. M. (1988). CDC definitions for
nosocomial infections, 1988. American Journal of Infection Control, 16(3), 128-140.
doi:10.1016/0196-6553(88)90053-3
Gautam, A., Ganu, S. S., Tegg, O. J., Andresen, D. N., Wilkins, B. H., & Schell, D. N. (2012).
Ventilator-associated pneumonia in a tertiary paediatric intensive care unit: A 1-year
prospective observational study. Critical Care & Resuscitation, 14(4), 283-289. Retrieved
from https://www.ncbi.nlm.nih.gov/pubmed/23230877.
Genaidy, A. M., Lemasters, G. K., Lockey, J., Succop, P., Deddens, J., Sobeih, T., & Dunning, K.
(2007). An epidemiological appraisal instrument – a tool for evaluation of epidemiological
studies. Ergonomics, 50(6), 920-960. doi:10.1080/00140130701237667
Giannini, M., Edwards, C., Rooks, T., Gilliam, C., Gipson, M., Owings, A., & Hakim, H. (2016). Is
monitoring parents, family, and visitors for hand hygiene compliance important? American
Journal of Infection Control, 44(6), S53. doi: 10.1016/j.ajic.2016.04.051
Gibson, M. F., Auld, V. E., Bryan, V. G., Coulson, V. S., Craig, V. J., & Glenny, V. A.-M. (2010).
A systematic review of oral assessment instruments: What can we recommend to practitioners
in children's and young people's cancer care? Cancer Nursing, 33(4), E1-E19.
doi:10.1097/NCC.0b013e3181cb40c0
Goutier, J. M., Holzmueller, C. G., Edwards, K. C., Klompas, M., Speck, K., & Berenholtz, S. M.
(2014). Strategies to enhance adoption of ventilator-associated pneumonia prevention
interventions: A systematic literature review. Infection Control and Hospital Epidemiology,
35(8), 998-1005. doi:10.1086/677152
Grohskopf, L. A., Sinkowitz-Cochran, R. L., Garrett, D. O., Sohn, A. H., Levine, G. L., Siegel, J. D.,
Pediatric Prevention, N. (2002). A national point-prevalence survey of pediatric intensive care
unit-acquired infections in the United States. Journal of Pediatrics, 140(4), 432-438.
doi:10.1067/mpd.2002.122499
Guess, R., Vaewpanich, J., Coss-Bu, J. A., Phongjitsiri, S., Kennedy, C., Starke, J., &
Thammasitboon, S. (2018). Risk factors for ventilator-associated events in a PICU. Pediatric
Critical Care Medicine, 19(1), e7-e13. doi:10.1097/PCC.0000000000001371
Gupta, A., Kapil, A., Kabra, S. K., Lodha, R., Sood, S., Dhawan, B., Sreenivas, V. (2014). Assessing
the impact of an educational intervention on ventilator-associated pneumonia in a pediatric
critical care unit. American Journal of Infection Control, 42(2), 111-115.
doi:10.1016/j.ajic.2013.09.026
Gupta, K., Gupta, V. K., Jayashree, M., & Singhi, S. (2012). Randomized controlled trial of
interrupted versus continuous sedative infusions in ventilated children. Pediatr Crit Care
Med, 13(2), 131-135. doi:10.1097/PCC.0b013e31820aba48
Gupta, R., & Rosen, D. (2016). Paediatric mechanical ventilation in the intensive care unit. BJA
Education, 16(12), 422-426. doi:10.1093/bjaed/mkw025
Gupta, S., Boville, B. M., Blanton, R., Lukasiewicz, G., Wincek, J., Bai, C., & Forbes, M. L. (2015).
A multicentered prospective analysis of diagnosis, risk factors, and outcomes associated with
175
pediatric ventilator-associated pneumonia. Pediatric Critical Care Medicine, 16(3), e65-73.
doi:10.1097/PCC.0000000000000338
Hamishehkar, H., Vahidinezhad, M., Mashayekhi, S. O., Asgharian, P., Hassankhani, H., &
Mahmoodpoor, A. (2014). Education alone is not enough in ventilator associated pneumonia
care bundle compliance. Journal of Research in Pharmacy Practise, 3(2), 51. Retrieved from
http://www.jrpp.net/
Hand Hygiene Australia. (2017a). Audit Tools. Retrieved from
http://www.hha.org.au/ForHealthcareWorkers/auditing.aspx#AuditTools
Hand Hygiene Australia. (2017b). National hand hygiene benchmark. Retrieved from
http://www.hha.org.au/LatestNationalData/national-data-for-2017/national-hand-hygiene-
benchmark.aspx
Harbarth, H. S., Pittet, A. D., Grady, A. L., Zawacki, A. A., Potter-Bynoe, A. G., Samore, A. M., &
Goldmann, A. D. (2002). Interventional study to evaluate the impact of an alcohol-based hand
gel in improving hand hygiene compliance. The Pediatric Infectious Disease Journal, 21(6),
489-495. doi:10.1097/00006454-200206000-00002
Harris, B. D., Hanson, C., Christy, C., Adams, T., Banks, A., Willis, T. S., & Maciejewski, M. L.
(2011). Strict hand hygiene and other practises shortened stays and cut costs and mortality in
a pediatric intensive care unit. Health Affairs (Project Hope), 30(9), 1751.
doi:10.1377/hlthaff.2010.1282
Hatachi, T., Tachibana, K., & Takeuchi, M. (2015). Incidences and influences of device-associated
healthcare-associated infections in a pediatric intensive care unit in Japan: A retrospective
surveillance study. Journal of Intensive Care, 3, 44. doi:10.1186/s40560-015-0111-6
Haustein, T., Gastmeier, P., Holmes, A., Lucet, J.-C., Shannon, R. P., Pittet, D., & Harbarth, S.
(2011). Use of benchmarking and public reporting for infection control in four high-income
countries. The Lancet Infectious Diseases, 11(6), 471-481. doi:10.1016/s1473-
3099(10)70315-7
Haut, C. (2015). Preventing pediatric ventilator-associated pneumonia. Nursing Critical Care, 10(6),
42-47. doi:10.1097/01.CCN.0000472845.62808.bb
Hayashi, Y., Morisawa, K., Klompas, M., Jones, M., Bandeshe, H., Boots, R., . Paterson, D. L. (2013).
Toward improved surveillance: the impact of ventilator-associated complications on length of
stay and antibiotic use in patients in intensive care units. Clinical Infectious Diseases, 56(4),
471-477. doi:10.1093/cid/cis926
Hebert, C., Flaherty, J., Smyer, J., Ding, J., & Mangino, J. E. (2017). Development and validation of
an automated ventilator-associated event electronic surveillance system: A report of a
successful implementation. AJIC: American Journal of Infection Control.
doi:10.1016/j.ajic.2017.09.006
Helder, O. K., Brug, J., Looman, C. W. N., van Goudoever, J. B., & Kornelisse, R. F. (2010). The
impact of an education program on hand hygiene compliance and nosocomial infection
incidence in an urban neonatal intensive care unit: An intervention study with before and after
comparison. International Journal of Nursing Studies, 47(10), 1245-1252.
doi:10.1016/j.ijnurstu.2010.03.005
Hellyer, T. P., Ewan, V., Wilson, P., & Simpson, A. J. (2016). The Intensive Care Society
recommended bundle of interventions for the prevention of ventilator-associated pneumonia.
Journal of the Intensive Care Society, 17(3), 238-243. doi:10.1177/1751143716644461
Heyland, D. K., Dhaliwal, R., Day, A., Jain, M., & Drover, J. (2004). Validation of the Canadian
clinical practise guidelines for nutrition support in mechanically ventilated, critically ill adult
patients: Results of a prospective observational study. Critical Care Medicine, 32(11), 2260-
2266. doi:10.1097/01.CCM.0000145581.54571.32
Hoog, M. d., Pickkers, P., Hoeven, J. G. v. d., Verlaat, C. W. M., Vet, N. J., Heesen, G. P., & Kox,
M. (2014). Randomized controlled trial of daily interruption of sedatives in critically ill
children. Paediatric Anaesthesia, 24(2), 151-156. doi:10.1111/pan.12245
176
Horan, T. C., Andrus, M., & Dudeck, M. A. (2008). CDC/NHSN surveillance definition of health
care–associated infection and criteria for specific types of infections in the acute care setting.
AJIC: American Journal of Infection Control, 36(5), 309-332. doi:10.1016/j.ajic.2008.03.002
Hsieh, H.-F., & Shannon, S. E. (2005). Three Approaches to Qualitative Content Analysis.
Qualitative Health Research, 15(9), 1277-1288. doi:10.1177/1049732305276687
IBM. (2016). SPSS Statistics for Window (Version 24.0). New York: IBM Corp.
IBM Knowledge Center. (2011). Data Cleaning. Retrieved from
https://www.ibm.com/support/knowledgecenter/en/SSLVQG_6.0.1/com.spss.ddl/dmgr_clea
ning.htm
Institute for Healhcare Improvement (IHI). (2015). Ventilator Bundle Checklist. Retrieved from
http://www.ihi.org/resources/Pages/Tools/VentilatorBundleChecklist.aspx
Iosifidis, E., Chochliourou, E., Violaki, A., Chorafa, E., Psachna, S., Roumpou, A., . .Roilides, E.
(2016). Evaluation of the new centers for disease control and prevention ventilator-associated
event module and criteria in critically ill children in Greece. Infection Control and Hospital
Epidemiology, 37(10), 1162-1166. doi:10.1017/ice.2016.135
Iosifidis, E., Stabouli, S., Tsolaki, A., Sigounas, V., Panagiotidou, E.-B., Sdougka, M., & Roilides,
E. (2015). Diagnosing ventilator-associated pneumonia in pediatric intensive care. AJIC:
American Journal of Infection Control, 43(4), 390-393. doi:10.1016/j.ajic.2015.01.004
Iregui, M., Ward, S., Sherman, G., Fraser, V. J., & Kollef, M. H. (2002). Clinical importance of delays
in the initiation of appropriate antibiotic treatment for ventilator-associated pneumonia. Chest,
122(1), 262-268. doi:10.1378/chest.122.1.262
Ismail, N., Darbyshire, A., & Thorburn, K. (2012). Ventilator-associated pneumonia (VAP): A UK
PICU experience. Archives of Disease in Childhood, 97(Suppl 1), A153-A154.
doi:10.1136/archdischild-2012-301885.361
Jaber, S., Amraoui, J., Lefrant, J. Y., Arich, C., Cohendy, R., Landreau, L., . . . Eledjam, J. J. (2006).
Clinical practise and risk factors for immediate complications of endotracheal intubation in
the intensive care unit: A prospective, multiple-center study. Critical Care Medicine, 34(9),
2355-2361. doi:10.1097/01.CCM.0000233879.58720.87
Jácomo, A. D. N., Carmona, F., Matsuno, A. K., Manso, P. H., & Carlotti, A. P. C. P. (2011). Effect
of oral hygiene with 0.12% chlorhexidine gluconate on the incidence of nosocomial
pneumonia in children undergoing cardiac surgery. Infection Control and Hospital
Epidemiology, 32(6), 591-596. doi:10.1086/660018
Jansen, A. C. M., van Aalst-Cohen, E. S., Hutten, B. A., Büller, H. R., Kastelein, J. J. P., & Prins, M.
H. (2005). Guidelines were developed for data collection from medical records for use in
retrospective analyses. Journal of Clinical Epidemiology, 58(3), 269-274.
doi:10.1016/j.jclinepi.2004.07.006
Jansson, M., Kääriäinen, M., & Kyngäs, H. (2013). Effectiveness of educational programmes in
preventing ventilator-associated pneumonia: A systematic review. Journal of Hospital
Infection, 84(3), 206-214. doi:10.1016/j.jhin.2013.04.009
Johnstone, L., Spence, D., & Koziol-McClain, J. (2010). Oral hygiene care in the pediatric intensive
care unit: Practise recommendations. Journal of Pediatric Nursing, 36(2), 85-96; quiz 97.
Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/20476510.
Jordan Garcia, I., Arriourtua, A. B., Torre, J. A., Anton, J. G., Vicente, J. C., & Gonzalez, C. T.
(2014). A national multicentre study on nosocomial infections in PICU. Anales de Pediatria
(Barcelona, Spain : 2003), 80(1), 28-33. doi:10.1016/j.anpedi.2010.09.010
Juneja, D., Singh, O., Javeri, Y., Arora, V., Dang, R., & Kaushal, A. (2011). Prevention and
management of ventilator-associated pneumonia: A survey on current practises by intensivists
practicing in the Indian subcontinent. Indian Journal of Anaesthesia, 55(2), 122-128.
doi:10.4103/0019-5049.79889
Kallet, R. H., & Quinn, T. E. (2005). The gastrointestinal tract and ventilator-associated pneumonia.
Respiratory Care, 50(7), 910-921; discussion 921-913. Retrieved from
https://www.ncbi.nlm.nih.gov/pubmed/15972112.
177
Karaaslan, A., Kepenekli Kadayifci, E., Atıcı, S., Sili, U., Soysal, A., Çulha, G., . Bakır, M. (2014).
Compliance of healthcare workers with hand hygiene practises in neonatal and pediatric
intensive care units: Overt observation. Interdisciplinary Perspectives on Infectious Diseases,
2014(3). doi:10.1155/2014/306478
Keszler, M. (2017). Mechanical ventilation strategies. Seminars in Fetal and Neonatal Medicine,
22(4), 267-274. doi:10.1016/j.siny.2017.06.003
Khemani, R. G., Markovitz, B. P., & Curley, M. A. Q. (2009). Characteristics of children intubated
and mechanically ventilated in 16 PICUs. Chest, 136(3), 765-771. doi:10.1378/chest.09-0207
Khine, H. H., Corddry, H. D., Kettrick, G. R., Martin, M. T., McCloskey, J. J., Rose, B. J., . Zagnoev,
C. M. (1997). Comparison of cuffed and uncuffed endotracheal tubes in young children during
general anesthesia. Anesthesiology, 86(3), 627-631. doi:10.1097/00000542-199703000-
00015
Khosim, R., Yaacob, N. A., Abdullah, N. H., Balakrishnan, S., Malar, V., Saad, N. A., & Suppian,
N. I. (2015). Reducing ventilator associated pneumonia (VAP) by implementing VAP bundle
checklist in critical care area. Journal of Microbiology, Immunology and Infection, 48(2), S93.
doi:10.1016/j.jmii.2015.02.330
Kim, M. K., Nam, E. Y., Na, S. H., Shin, M. J., Lee, H. S., Kim, N. H., . .Kim, H. B. (2015).
Discrepancy in perceptions regarding patient participation in hand hygiene between patients
and health care workers. American Journal of Infection Control, 43(5), 510-515.
doi:10.1016/j.ajic.2015.01.018
King, G., & Zeng, L. (2001). Logistic regression in rare events data. Political analysis, 9(2), 137-163.
Klein Klouwenberg, P. M., van Mourik, M. S., Ong, D. S., Horn, J., Schultz, M. J., Cremer, O. L., .
Consortium, M. (2014). Electronic implementation of a novel surveillance paradigm for
ventilator-associated events. Feasibility and validation. American Journal of Respiratory
Critical Care Medicine, 189(8), 947-955. doi:10.1164/rccm.201307-1376OC
Klein Klouwenberg, P. M., van Mourik, M. S. M., Ong, D. S. Y., Horn, J., Schultz, M. J., Cremer, O.
L., & Bonten, M. J. M. (2013). Validation of a novel surveillance paradigm for ventilator-
associated events. Critical Care, 17(Suppl 4), P1-P1. doi:10.1186/cc12902
Klompas, M. (2010). Interobserver variability in ventilator-associated pneumonia surveillance.
American Journal of Infection Control, 38(3), 237-239. doi:10.1016/j.ajic.2009.10.003
Klompas, M. (2012). Advancing the science of ventilator-associated pneumonia surveillance. Critical
Care, 16(5), 165. doi:10.1186/cc11656
Klompas, M. (2013a). Complications of Mechanical Ventilation — The CDC's New Surveillance
Paradigm. The New England Journal of Medicine, 368(16), 1472-1475.
doi:10.1056/NEJMp1300633
Klompas, M. (2013b). Ventilator-associated events surveillance: A patient safety opportunity.
Current Opinion in Critical Care, 19(5), 424-431. doi:10.1097/MCC.0b013e3283636bc9
Klompas, M. (2015). Potential Strategies to Prevent Ventilator- associated Events. In American
Journal of Respiratory and Critical Care Medicine. (Vol. 192, pp. 1420-1430).
Klompas, M., Anderson, D., Trick, W., Babcock, H., Kerlin, M. P., Li, L., . Epicenters, C. D. C. P.
(2015). The preventability of ventilator-associated events. The CDC Prevention Epicenters
Wake Up and Breathe Collaborative. American Journal of Respiratory and Critical Care
Medicine, 191(3), 292-301. doi:10.1164/rccm.201407-1394OC
Klompas, M., Branson, R., Eichenwald, E. C., Greene, L. R., Howell, M. D., Lee, G., . Society for
Healthcare Epidemiology of, A. (2014). Strategies to prevent ventilator-associated pneumonia
in acute care hospitals: 2014 update. Infection Control & Hospital Epidemiology, 35(8), 915-
936. doi:10.1086/677144
Klompas, M., Khan, Y., Kleinman, K., Evans, R. S., Lloyd, J. F., Stevenson, K., . .Program, C. D. C.
P. E. (2011). Multicenter evaluation of a novel surveillance paradigm for complications of
mechanical ventilation. PLOS One, 6(3), e18062. doi:10.1371/journal.pone.0018062
178
Klompas, M., Kleinman, K., Khan, Y., Evans, R. S., Lloyd, J. F., Stevenson, K., . for the, C. D. C. P.
E. P. (2012). Rapid and reproducible surveillance for ventilator-associated pneumonia.
Clinical Infectious Diseases, 54(3), 370-377. doi:10.1093/cid/cir832
Klompas, M., Kleinman, K., & Murphy, M. V. (2014). Descriptive epidemiology and attributable
morbidity of ventilator-associated events. Infection Control & Hospital Epidemiology, 35(5),
502-510. doi:10.1086/675834
Klompas, M., Li, L., Kleinman, K., Szumita, P. M., & Massaro, A. F. (2016). Associations between
ventilator bundle components and outcomes. JAMA Internal Medicine, 176(9), 1277-1283.
doi:10.1001/jamainternmed.2016.2427
Klompas, M., Magill, S., Robicsek, A., Strymish, J. M., Kleinman, K., Evans, R. S., . Program, C. D.
C. P. E. (2012). Objective surveillance definitions for ventilator-associated pneumonia.
Critical Care Medicine, 40(12), 3154-3161. doi:10.1097/CCM.0b013e318260c6d9
Klompas, M., Speck, K., Howell, M. D., Greene, L. R., & Berenholtz, S. M. (2014). Reappraisal of
routine oral care with chlorhexidine gluconate for patients receiving mechanical ventilation:
Systematic review and meta-analysis. JAMA Internal Medicine, 174(5), 751-761.
doi:10.1001/jamainternmed.2014.359
Kobayashi, H., Pian-Smith, M., Sato, M., Sawa, R., Takeshita, T., & Raemer, D. (2006). A cross-
cultural survey of residents’ perceived barriers in questioning/challenging authority. BMJ
Quality & Safety, 15(4), 277-283.
Koff, M. D., Corwin, H. L., Beach, M. L., Surgenor, S. D., & Loftus, R. W. (2011). Reduction in
ventilator associated pneumonia in a mixed intensive care unit after initiation of a novel hand
hygiene program. Journal of Critical Care, 26(5), 489-495. doi:10.1016/j.jcrc.2010.12.013
Koff, M. D., Loftus, R. W., Burchman, C. C., Schwartzman, J. D., Read, M. E., Henry, E. S., &
Beach, M. L. (2009). Reduction in intraoperative bacterial contamination of peripheral
intravenous tubing through the use of a novel device. Anesthesiology, 110(5), 978-985.
doi:10.1097/ALN.0b013e3181a06ec3
Kollef, M. H. (2004). Prevention of hospital-associated pneumonia and ventilator-associated
pneumonia. Critical Care Medicine, 32(6), 1396-1405.
doi:10.1097/01.CCM.0000128569.09113.FB
Kollef, M. H. (2011). Prevention of Nosocomial Pneumonia in the Intensive Care Unit: Beyond the
use of bundles. Surgical Infections, 12(3), 211-220. doi:10.1089/sur.2010.060
Kozlowski, D. (2012). Physical interventions to interrupt or reduce the spread of respiratory viruses.
International Journal of Evidence‐Based Healthcare, 10(2), 159-161. doi:10.1111/j.1744-
1609.2012.00268.x
Kunzmann, H., Dimitriades, K., Morrow, B., & Argent, A. (2016). Reducing paediatric ventilator-
associated pneumonia--a South African challenge! Southern African Journal of Critical Care,
32(1), 17. doi:10.7196/SAJCC.2016.v32i1.243
Kuo, D. Z., Houtrow, A. J., Arango, P., Kuhlthau, K. A., Simmons, J. M., & Neff, J. M. (2012).
Family-Centered Care: Current applications and future directions in pediatric health care.
Maternal and Child Health Journal, 16(2), 297-305. doi:10.1007/s10995-011-0751-7
Kusahara, D. M., Enz Cda, C., Avelar, A. F., Peterlini, M. A., & Pedreira Mda, L. (2014). Risk factors
for ventilator-associated pneumonia in infants and children: A cross-sectional cohort study.
American Journal of Critical Care, 23(6), 469-476. doi:10.4037/ajcc2014127
Kusahara, D. M., Peterlini, M. A., & Pedreira, M. L. (2012). Oral care with 0.12% chlorhexidine for
the prevention of ventilator-associated pneumonia in critically ill children: Randomised,
controlled and double blind trial. International Journal of Nursing Studies, 49(11), 1354-
1363. doi:10.1016/j.ijnurstu.2012.06.005
Labeau, S., Bleiman, M., Rello, J., Vandijck, D., Claes, B., Blot, S., & Grp, E. S. (2009). Intensive
care nurses ' knowledge and management of tracheal tube cuffs. Intensive Care Medicine, 35,
203-203. Retrieved from http://icmjournal.esicm.org/index.html.
179
Labeau, S., Rello, J., Dimopoulos, G., Lipman, J., Sarikaya, A., Ozturk, C., . Blot, S. (2016). The
value of E-Learning for the prevention of Healthcare-Associated Infections. Infection Control
& Hospital Epidemiology, 37(9), 1052-1059. doi:10.1017/ice.2016.107
Lawrence, P., & Fulbrook, P. (2012). Effect of feedback on ventilator care bundle compliance: Before
and after study. Nursing in Critical Care, 17(6), 293-301. doi:10.1111/j.1478-
5153.2012.00519.x
Leape, L. L., Berwick, D., Clancy, C., Conway, J., Gluck, P., Guest, J., . . Lucian Leape Institute at
the National Patient Safety, F. (2009). Transforming healthcare: A safety imperative. Quality
and Safety in Health Care, 18(6), 424-428. doi:10.1136/qshc.2009.036954
Leong, S., Bass, P., Worth, L., Miley, J., Klintworth, G., Spilsbury, S., & Watson, K. (2017). ‘It's all
about technique’: Monitoring compliance with standardised aseptic technique practises at a
large Australian health service. Infection, Disease & Health, 22, S1. doi:
10.1016/j.idh.2017.09.005
Levchenko, A. I., Hufton, G. C., Boscart, V. M., & Fernie, G. R. (2010). Embedded System for
Hygiene Compliance Monitoring. Automation Science and Engineering, IEEE Transactions
on, 7(3), 695-698. doi:10.1109/TASE.2010.2044882
Lewis, S. C., Li, L., Murphy, M. V., Klompas, M., Epictr, C. D. C. P., & Epicenters, C. D. C. P.
(2014). Risk factors for centilator-associated events: A case-control multivariable analysis.
Critical Care Medicine, 42(8), 1839-1848. doi:10.1097/CCM.0000000000000338
Liu, B., Li, S. Q., Zhang, S. M., Xu, P., Zhang, X., Zhang, Y. H., . . . Zhang, W. H. (2013). Risk
factors of ventilator-associated pneumonia in pediatric intensive care unit: A systematic
review and meta-analysis. Journal of Thoracic Diseases, 5(4), 525-531.
doi:10.3978/j.issn.2072-1439.2013.08.31
Longtin, Y., Sax, H., Allegranzi, B., Hugonnet, S., & Pittet, D. (2009). Patients’ Beliefs and
Perceptions of Their Participation to Increase Healthcare Worker Compliance with Hand
Hygiene. Infection Control and Hospital Epidemiology, 30(9), 830-839. doi:10.1086/599118
Lopriore, E., Markhorst, D., & Gemke, R. (2002). Ventilator-associated pneumonia and upper airway
colonisation with Gram negative bacilli: The role of stress ulcer prophylaxis in children.
Intensive Care Medicine, 28(6), 763-767. doi:10.1007/s00134-002-1289-3
Lorente, L., Lecuona, M., Martín, M. M., García, C., Mora, M. L., & Sierra, A. (2005). Ventilator-
associated pneumonia using a closed versus an open tracheal suction system. Critical Care
Medicine, 33(1), 115-119. doi:10.1097/01.Ccm.0000150267.40396.90
Lu, C.-T., Shin, T.-P., & Ding, L.-W. (2015). Reducing ventilator-associated pneumonia in intensive
care unit of a community hospital: Impact of implementing VAP bundle. Journal of
Microbiology, Immunology and Infection, 48(2), S89-S90. doi:10.1016/j.jmii.2015.02.316
Lutmer, J. E., & Brilli, R. J. (2016). Unanswered Questions and Consternation: The Ventilator-
Associated Pneumonia Diagnostic Challenge Continues. Pediatric Critical Care Medicine,
17(2), 175-176. doi:10.1097/PCC.0000000000000598
Magill, S. S., Klompas, M., Balk, R., Burns, S. M., Deutschman, C. S., Diekema, D., . . Lipsett, P.
(2013). Developing a new, national approach to surveillance for ventilator-associated events.
Critical Care Medicine, 41(11), 2467-2475. doi:10.1097/CCM.0b013e3182a262db
Magill, S. S., Rhodes, B., & Klompas, M. (2014). Improving ventilator-associated event surveillance
in the National Healthcare Safety Network and addressing knowledge gaps: Update and
review. Current Opinion in Infectious Disease, 27(4), 394-400.
doi:10.1097/QCO.0000000000000083
Marcia Regina Eches, P., Perugini, V. H., Figueira, F. D., Louise Marina Silva, F., Diniz, J. J., Daniela
Lucio dos, S., . . Capobiango, J. D. (2015). Impact of the ventilator bundle on ventilator-
associated pneumonia (VAP) rates in a pediatric intensive care unit in Londrina-PR. Semina:
Ciências Biológicas e da Saúde, 36(1Supl), 259-266. doi:10.5433/1679-
0367.2015v36n1Suplp259
Martin, M., Zingg, W., Hansen, S., Gastmeier, P., Wu, A. W., Pittet, D., & Dettenkofer, M. (2013).
Public reporting of healthcare-associated infection data in Europe. What are the views of
180
infection prevention opinion leaders? Journal of Hospital Infection, 83(2), 94-98.
doi:10.1016/j.jhin.2012.10.010
McCambridge, J., Witton, J., & Elbourne, D. R. (2014). Systematic review of the Hawthorne effect:
new concepts are needed to study research participation effects. Journal of Clinical
Epidemiology, 67(3), 267-277. doi:doi:10.1016/j.jclinepi.2013.08.015
McGuckin, M., Storr, J., Longtin, Y., Allegranzi, B., & Pittet, D. (2011). Patient empowerment and
multimodal hand hygiene promotion: A Win-Win strategy. American Journal of Medical
Quality, 26(1), 10-17. doi:10.1177/1062860610373138
MEDCALC easy-to-use statistical software. (2017). Retrieved from
https://www.medcalc.org/calc/diagnostic_test.php
Meert, K. L., Clark, J., & Eggly, S. (2013). Family-Centered Care in the pediatric intensive care unit.
Pediatric Clinics of North America, 60(3), 761. doi:10.1016/j.pcl.2013.02.011
Memela, M. E., & Gopalan, P. D. (2014). Variations in endotracheal tube cuff pressure: Is 8-hourly
monitoring enough? Southern African Journal of Critical Care, 30(2), 35.
doi:10.7196/SAJCC.159
Mhanna, M. J. (2016). Ventilator-associated events in neonates and children: A single definition for
a heterogeneous population. Critical Care Medicine, 44(1), 233-234.
doi:10.1097/CCM.0000000000001443
Mietto, C., Pinciroli, R., Patel, N., & Berra, L. (2013). Ventilator associated pneumonia: Evolving
definitions and preventive strategies. Respiratory Care, 58(6), 990-1007.
doi:10.4187/respcare.02380
Mohd Ali, N. A., Jauncey-Cooke, J., & Bogossian, F. (2019). Ventilator-associated events in children:
A review of literature. Australian Critical Care, 32(1), 55-62. doi:10.1016/j.aucc.2018.11.063
Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & Group, P. (2009). Preferred reporting items for
systematic reviews and meta-analyses: The PRISMA statement. Journal of Clinical
Epidemiology, 62(10), 1006-1012. doi:10.1016/j.jclinepi.2009.06.005
Morinec, J., Iacaboni, J., & McNett, M. (2012). Risk factors and interventions for ventilator-
associated pneumonia in pediatric patients. Journal of Pediatric Nursing, 27(5), 435-442.
doi:10.1016/j.pedn.2012.03.027
Morris, A. C., Hay, A. W., Swann, D. G., Everingham, K., McCulloch, C., McNulty, J., . . . Walsh,
T. S. (2011). Reducing ventilator-associated pneumonia in intensive care: impact of
implementing a care bundle. Critical Care Medicine, 39(10), 2218-2224.
doi:10.1097/CCM.0b013e3182227d52
Morrow, B. M., & Argent, A. C. (2008). A comprehensive review of pediatric endotracheal
suctioning: Effects, indications, and clinical practise. Pediatric Critical Care Medicine, 9(5),
465-477. doi:10.1097/PCC.0b013e31818499cc
Morrow, B. M., Argent, A. C., Jeena, P. M., & Green, R. J. (2009). Guideline for the diagnosis,
prevention and treatment of paediatric ventilator-associated pneumonia. South African
Medical Journal, 99(4 Pt 2), 255-267.
Morrow, B. M., Mowzer, R., Pitcher, R., & Argent, A. C. (2012). Investigation into the effect of
closed-system suctioning on the frequency of pediatric ventilator-associated pneumonia in a
developing country. Pediatric Critical Care Medicine, 13(1), e25-32.
doi:10.1097/PCC.0b013e31820ac0a2
Mourani, P., & Sontag, M. (2017). Ventilator-associated pneumonia in critically ill children: A new
paradigm. Pediatric Clinics of North America, 64(5), 1039-+. doi:10.1016/j.pcl.2017.06.005
Muething, S. E., Goudie, A., Schoettker, P. J., Donnelly, L. F., Goodfriend, M. A., Bracke, T. M., . .
Kotagal, U. R. (2012). Quality improvement initiative to reduce serious safety events and
improve patient safety culture. Pediatrics, 130(2), e423. doi:10.1542/peds.2011-3566
Munro, C., Grap, M. J., Elswick, R. K., Jr., McKinney, J., Sessler, C. N., & Hummel, R. S., 3rd.
(2006). Oral health status and development of ventilator-associated pneumonia: A descriptive
study. American Journal of Critical Care, 15(5), 453-460. Retrieved from
https://www.ncbi.nlm.nih.gov/pubmed/16926366.
181
Munro, N., & Ruggiero, M. (2014). Ventilator-associated pneumonia bundle: Reconstruction for best
care. AACN Advanced Critical Care, 25(2), 163-175; 176-168.
doi:10.1097/NCI.0000000000000019
Muscedere, J., Dodek, P., Keenan, S., Fowler, R., Cook, D., Heyland, D., . . the Canadian Critical
Care Trials, G. (2008). Comprehensive evidence-based clinical practise guidelines for
ventilator-associated pneumonia: Diagnosis and treatment. Journal of Critical Care, 23(1),
138-147. doi:10.1016/j.jcrc.2007.12.008
Muscedere, J., Sinuff, T., Heyland, D. K., Dodek, P. M., Keenan, S. P., Wood, G., . . . Canadian
Critical Care Trials, G. (2013). The clinical impact and preventability of ventilator-associated
conditions in critically ill patients who are mechanically ventilated. Chest, 144(5), 1453-1460.
doi:10.1378/chest.13-0853
Muscedere, J. G., Day, A., & Heyland, D. K. (2010). Mortality, attributable mortality, and clinical
events as end points for clinical trials of ventilator-associated pneumonia and hospital-
acquired pneumonia. Clinical Infectious Disease, 51 Suppl 1(S1), S120-125.
doi:10.1086/653060
Muszynski, J. A., Sartori, J., Steele, L., Frost, R., Wang, W., Khan, N., . .Ayad, O. (2013).
Multidisciplinary quality improvement initiative to reduce ventilator-associated
tracheobronchitis in the PICU. Pediatric Critical Care Medicine, 14.
doi:10.1097/PCC.0b013e31828a897f
Nair, G. B., & Niederman, M. S. (2015). Ventilator-associated pneumonia:Present understanding and
ongoing debates. Intensive Care Medicine, 41(1), 34-48. doi:10.1007/s00134-014-3564-5
Narayanan, A., Dixon, G., Chalkley, S., Ray, S., & Brierley, J. (2016). Ventilator-associated
pneumonia in children: Comparing the plethora of surveillance definitions. Journal of
Hospital Infection, 94(2), 163-164. doi:10.1016/j.jhin.2016.05.006
Navoa-Ng, J. A., Berba, R., Galapia, Y. A., Rosenthal, V. D., Villanueva, V. D., Tolentino, M. C. V.,
Mantaring, J. B. V. (2011). Device-associated infections rates in adult, pediatric, and neonatal
intensive care units of hospitals in the Philippines: International Nosocomial Infection Control
Consortium (INICC) findings. AJIC: American Journal of Infection Control, 39(7), 548-554.
doi:10.1016/j.ajic.2010.10.018
Neto, A. S., Simonis, F. D., Barbas, C. S., Biehl, M., Determann, R. M., Elmer, J., . . Linko, R. (2015).
Lung-protective ventilation with low tidal volumes and the occurrence of pulmonary
complications in patients without acute respiratory distress syndrome: A systematic review
and individual patient data analysis. Critical Care Medicine, 43(10), 2155-2163. doi:
10.1097/CCM.0000000000001189
Niel-Weise, B. S., Gastmeier, P., Kola, A., Vonberg, R. P., Wille, J. C., van den Broek, P. J., & Bed
Head Elevation Study, G. (2011). An evidence-based recommendation on bed head elevation
for mechanically ventilated patients. Critical Care, 15(2), R111. doi:10.1186/cc10135
Nix, S., Smith, M., & Vicenzino, B. (2010). Prevalence of hallux valgus in the general population: a
systematic review and meta-analysis. Journal of Foot and Ankle Research, 3(1), 21.
doi:10.1186/1757-1146-3-21
Obeid, A., Naous, A., Naja, Z., & Naja, A. S. (2014). Preventing ventilator associated pneumonia
(VAP) in a pediatric intensive care unit using a modified ventilator associated pneumonia
bundle: Pre-interventional and post-interventional trial. Research Journal of Medical
Sciences, 8(1), 13-19. doi:10.3923/rjmsci.2014.13.19
Okuyama, A., Wagner, C., & Bijnen, A. B. (2014). Speaking up for patient safety by hospital-based
health care professionals: A literature review. BMC Health Services Research, 14,
urn:issn:1472-6963. doi:10.1186/1472-6963-14-61
Ostermann, M. E., Keenan, S. P., Seiferling, R. A., & Sibbald, W. J. (2000). Sedation in the intensive
care unit: a systematic review. JAMA, 283(11), 1451-1459. doi: 10.1001/jama.283.11.1451
Owen, E. B., Woods, C. R., O'Flynn, J. A., Boone, M. C., Calhoun, A. W., & Montgomery, V. L.
(2016). A Bedside Decision Tree for Use of Saline With Endotracheal Tube Suctioning in
Children. Critical Care Nurse, 36(1), e1. doi:10.4037/ccn2016358
182
Paju, S., & Scannapieco, F. (2007). Oral biofilms, periodontitis, and pulmonary infections. Oral
diseases, 13(6), 508-512. doi: 10.1111/j.1601-0825.2007.01410a.x
Pan, S. C., Tien, K. L., Hung, I. C., Lin, Y. J., Yang, Y. L., Yang, M. C., . .Chen, Y. C. (2013). Patient
empowerment in a hand hygiene program: Differing points of view between patients/family
members and health care workers in Asian culture. American Journal of Infection Control,
41(11), 979-983. doi:10.1016/j.ajic.2013.02.008
Parsons, H. M. (1974). What Happened at Hawthorne? Science, 183(4128), 922-932.
doi:10.1126/science.183.4128.922
Patria, M. F., Chidini, G., Ughi, L., Montani, C., Prandi, E., Galeone, C., . .Esposito, S. (2013).
Ventilator-associated pneumonia in an Italian pediatric intensive care unit: A prospective
study. World Journal of Pediatrics, 9(4), 365-368. doi:10.1007/s12519-013-0444-y
Patrick, S. W., Kawai, A. T., Kleinman, K., Jin, R., Vaz, L., Gay, C., . . Lee, G. M. (2014). Health
care-associated infections among critically ill children in the US, 2007–2012. Pediatrics, 134.
doi:10.1542/peds.2014-0613
Pearson, L. S., & Hutton, J. L. (2002). A controlled trial to compare the ability of foam swabs and
toothbrushes to remove dental plaque. Journal of Advanced Nursing, 39(5), 480-489.
doi:10.1046/j.1365-2648.2002.02313.x
Phongjitsiri, S., Coss-Bu, J., Kennedy, C., Silva, J., Starke, J., Graf, J., & Thammasitboon, S. (2015).
The Centers for Disease Control and Prevention's new definitions for complications of
mechanical ventilation shift the focus of quality surveillance and predict clinical outcomes in
a PICU. Critical Care Medicine, 43(11), 2446-2451. doi:10.1097/CCM.0000000000001261
Piper, L. E. (2011). The ethical leadership challenge: Creating a culture of patient- and family-
centered care in the hospital setting. Health Care Manager, 30(2), 125-132.
doi:10.1097/HCM.0b013e318216efb9
Pittet, D. (2010). Reducing ventilator-associated pneumonia: When process control allows outcome
improvement and even benchmarking. Critical Care Medicine, 38(3), 983-984.
doi:10.1097/CCM.0b013e3181c8fd0c
Pittet, D., Panesar, S. S., Wilson, K., Longtin, Y., Morris, T., Allan, V., . . . Donaldson, L. (2011).
Involving the patient to ask about hospital hand hygiene: A National Patient Safety Agency
feasibility study. Journal of Hospital Infection, 77(4), 299-303.
doi:10.1016/j.jhin.2010.10.013
Porta, M., Porta, M. S., Hernâan, M., Last, J. M., Greenland, S., Hernán, M., & Silva, I. d. S. (2014).
A dictionary of epidemiology (6th ed.). New York: New York : Oxford University Press.
Prakash, V., Parameswaran, N., & Biswal, N. (2016). Early versus late enteral feeding in critically ill
children: A randomized controlled trial. Intensive Care Medicine, 42(3), 481-482.
doi:10.1007/s00134-015-4176-4
Qualtrics. (2016). Qualtrics Software. Provo, UT, USA. Retrieved from http://www.qualtrics.com
Queensland Children's Hospital Paediatric Intensive Care Unit. (2016). Ventilator Associated
Pneumonia Prevention Strategies VAP BUNDLE [Poster]. In. Brisbane, Australia: Ivy Chang.
Queensland Children’s Hospital Paediatric Intensive Care Unit. (2015). Oral Hygiene Resource
[Powerpoint slides]. Unpublished manuscript. Lady Cilento Children's Hospital, Brisbane,
Australia.
Rabelo, M. A., Bezerra Neto, A. M., Loibman, S. O., Lima, J. L., Ferreira, E. L., Leal, N. C., &
Maciel, M. A. (2014). The occurrence and dissemination of methicillin and vancomycin-
resistant Staphylococcus in samples from patients and health professionals of a university
hospital in Recife, State of Pernambuco, Brazil. Revista da Sociedade Brasileira de Medicina
Tropical, 47(4), 437-446. doi: 10.1590/0037-8682-0071-2014
Raoof, S., Baumann, M. H., Critical Care Societies Collaborative, & The Society of Critical Care
Medicine. (2014). Ventilator-associated events: The new definition. American Joural of
Critital Care, 23(1), 7-9. doi:10.4037/ajcc2014469
183
Raoof, S., Baumann, M. H., & Critical Care Society Collaborative. (2014). An official multi-society
statement ventilator-associated events: The new definition. Chest, 145(1), 10-12.
doi:10.1378/chest.13-2731
Rasslan, O., Seliem, Z. S., Ghazi, I. A., El Sabour, M. A., El Kholy, A. A., Sadeq, F. M., . . . Rosenthal,
V. D. (2012). Device-associated infection rates in adult and pediatric intensive care units of
hospitals in Egypt. International Nosocomial Infection Control Consortium (INICC) findings.
Journal of Infection and Public Health, 5(6), 394-402. doi:10.1016/j.jiph.2012.07.002
Rebmann, T., & Greene, L. R. (2010). Preventing ventilator-associated pneumonia: An executive
summary of the Association for Professionals in Infection Control and Epidemiology, Inc,
Elimination Guide. AJIC: American Journal of Infection Control, 38(8), 647-649.
doi:10.1016/j.ajic.2010.08.004
Reime, M. H., Harris, A., Aksnes, J., & Mikkelsen, J. (2008). The most successful method in teaching
nursing students infection control – E-learning or lecture? Nurse Education Today, 28(7), 798-
806. doi:10.1016/j.nedt.2008.03.005
Rello, J., Afonso, E., Lisboa, T., Ricart, M., Balsera, B., Rovira, A., . . . Investigators, F. P. (2013).
A care bundle approach for prevention of ventilator-associated pneumonia. Clinical
Microbiology and Infection, 19(4), 363-369. doi:10.1111/j.1469-0691.2012.03808.x
Rello, J., Lode, H., Cornaglia, G., Masterton, R., & Contributors, V. A. P. C. B. (2010). A European
care bundle for prevention of ventilator-associated pneumonia. Intensive Care Medicine,
36(5), 773-780. doi:10.1007/s00134-010-1841-5
Rello, J., Lorente, C., Bodí, M., Diaz, E., Ricart, M., & Kollef, M. H. (2002). Why do physicians not
follow evidence-based guidelines for preventing ventilator-associated pneumonia? A survey
based on the opinions of an international panel of intensivists. Chest, 122(2), 656-661.
doi:10.1378/chest.122.2.656
Resar, R., Griffin, F., Haraden, C., & Nolan, T. (2014). Using care bundles to improve health care
quality. IHI innovation series white paper. Cambridge, MA: Institute for Healthcare
Improvement; 2012. Retrived from www.IHI.org
Resar, R., Pronovost, P., Haraden, C., Simmonds, T., Rainey, T., & Nolan, T. (2005). Using a bundle
approach to improve ventilator care processes and reduce ventilator-associated pneumonia.
The Joint Commission Journal on Quality and Patient Safety, 31(5), 243-248.
doi:10.1016/s1553-7250(05)31031-2
Restrepo, M. I., Anzueto, A., Arroliga, A. C., Afessa, B., Atkinson, M. J., Ho, N. J., . . Kollef, M. H.
(2010). Economic burden of ventilator-associated pneumonia based on total resource
utilization. Infection Control & Hospital Epidemiology, 31(5), 509-515. doi:10.1086/651669
Roeleveld, P. P., Guijt, D., Kuijper, E. J., Hazekamp, M. G., de Wilde, R. B., & de Jonge, E. (2011).
Ventilator-associated pneumonia in children after cardiac surgery in The Netherlands.
Intensive Care Medicine, 37(10), 1656-1663. doi:10.1007/s00134-011-2349-3
Rosen, P., Stenger, E., Bochkoris, M., Hannon, M. J., & Kwoh, C. K. (2009). Family-centered
multidisciplinary rounds enhance the team approach in pediatrics. Pediatrics, 123(4), e603.
doi:10.1542/peds.2008-2238
Rosenthal, V. D., Alvarez-Moreno, C., Villamil-Gomez, W., Singh, S., Ramachandran, B., Navoa-
Ng, J. A., . . Kuyucu, N. (2012). Effectiveness of a multidimensional approach to reduce
ventilator-associated pneumonia in pediatric intensive care units of 5 developing countries:
International Nosocomial Infection Control Consortium findings. American Journal of
Infection Control, 40(6), 497-501. doi:10.1016/j.ajic.2011.08.005
Rosenthal, V. D., Bijie, H., Maki, D. G., Mehta, Y., Apisarnthanarak, A., Medeiros, E. A., . .
members, I. (2012). International Nosocomial Infection Control Consortium (INICC) report,
data summary of 36 countries, for 2004-2009. American Journal of Infection Control, 40(5),
396-407. doi:10.1016/j.ajic.2011.05.020
Rosenthal, V. D., Guzman, S., Pezzotto, S. M., & Crnich, C. J. (2003). Effect of an infection control
program using education and performance feedback on rates of intravascular device-
184
associated bloodstream infections in intensive care units in Argentina. American Journal of
Infection Control, 31(7), 405-409. doi:10.1067/mic.2003.52
Rosenthal, V. D., Guzman, S., & Safdar, N. (2005). Reduction in nosocomial infection with improved
hand hygiene in intensive care units of a tertiary care hospital in Argentina. American Journal
of Infection Control, 33(7), 392-397. doi:10.1016/j.ajic.2004.08.009
Sachdev, R. A., Chugh, R. K., Raghunathan, R. V., Gupta, R. D., Wattal, R. C., & Menon, R. G.
(2013). Diagnosis of bacterial ventilator–associated pneumonia in children: Reproducibility
of Blind Bronchial Sampling. Pediatric Critical Care Medicine, 14(1), e1-e7.
doi:10.1097/PCC.0b013e318260129d
Safdar, N., Crnich, C. J., & Maki, D. G. (2005). The pathogenesis of ventilator-associated pneumonia:
Its relevance to developing effective strategies for prevention. Respiratory Care, 50(6), 725-
739; discussion 739-741. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/15913465.
Samransamruajkit, R., Jirapaiboonsuk, S., Siritantiwat, S., Tungsrijitdee, O., Deerojanawong, J.,
Sritippayawan, S., & Prapphal, N. (2010). Effect of frequency of ventilator circuit changes (3
vs 7 days) on the rate of ventilator-associated pneumonia in PICU. Journal of Critical Care,
25(1), 56-61. doi:10.1016/j.jcrc.2009.03.005
Samuel, R., Shuen, A., Dendle, C., Kotsanas, D., Scott, C., & Stuart, R. L. (2012). Hierarchy and
Hand Hygiene: Would Medical Students Speak Up to Prevent Hospital-Acquired Infection?
Infection Control and Hospital Epidemiology, 33(8), 861-863. doi:10.1086/666634
Santschi, G. M., Randolph, C. A., Rimensberger, C. P., & Jouvet, C. P. (2013). Mechanical
Ventilation Strategies in Children With Acute Lung Injury: A Survey on Stated Practise
Pattern. Pediatric Critical Care Medicine, 14(7), e332-e337.
doi:10.1097/PCC.0b013e31828a89a2
Scannapieco, F. A., Yu, J., Raghavendran, K., Vacanti, A., Owens, S. I., Wood, K., & Mylotte, J. M.
(2009). A randomized trial of chlorhexidine gluconate on oral bacterial pathogens in
mechanically ventilated patients. Critical Care, 13(4), R117. doi:10.1186/cc7967
Schults, J., Mitchell, M. L., Cooke, M., & Schibler, A. (2018). Efficacy and safety of normal saline
instillation and paediatric endotracheal suction: An integrative review. Australian Critical
Care, 31(1), 3-9. doi:10.1016/j.aucc.2017.02.069
Schwappach, D. L. B., Frank, O., & Davis, R. E. (2013). A vignette study to examine health care
professionals'attitudes towards patient involvement in error prevention. Journal of Evaluation
in Clinical Practise, 19(5), 840-848. doi:10.1111/j.1365-2753.2012.01861.x
Scoble, M. K., Copnell, B., Taylor, A., Kinney, S., & Shann, F. (2001). Effect of reusing suction
catheters on the occurrence of pneumonia in children. Heart & Lung, 30(3), 225-233.
Septimus, E., Green, L., & Klompas, M. (2015). Ventilator-associated events: A broader perspective.
Critical Care Medicine, 43(2), e59-61. doi:10.1097/CCM.0000000000000638
Sergeant, E. (2017). Epitools epidemiological calculators. Retrieved from
http://epitools.ausvet.com.au.
Shmilev, T. I., & Yankov, I. V. (2012). Ventilator-associated pneumonias in children (II)--
prophylaxis and treatment. Folia Medica (Plovdiv), 54(1), 12-18. Retrieved from
https://www.ncbi.nlm.nih.gov/pubmed/22908825.
Sickbert-Bennett, E. E., DiBiase, L. M., Willis, T. M. S., Wolak, E. S., Weber, D. J., & Rutala, W.
A. (2016). Reduction of healthcare-associated infections by exceeding high compliance with
hand hygiene practises.(DISPATCHES). Emerging Infectious Diseases, 22(9), 1628.
doi:10.3201/eid2209.151440
Siegel, J. D., Rhinehart, E., Jackson, M., Chiarello, L., & Health Care Infection Control Practises
Advisory, C. (2007). 2007 Guideline for isolation precautions: Preventing transmission of
infectious agents in health care settings. American Journal of Infection Control, 35(10 Suppl
2), S65-164. doi:10.1016/j.ajic.2007.10.007
Singh, S., Kumar, R. K., Sundaram, K. R., Kanjilal, B., & Nair, P. (2012). Improving outcomes and
reducing costs by modular training in infection control in a resource-limited setting.
International Journal for Quality in Health Care, 24(6), 641-648. doi:10.1093/intqhc/mzs059
185
Sinuff, T., Muscedere, J., Cook, D. J., Dodek, P. M., Anderson, W., Keenan, S. P., . . Canadian
Critical Care Trials, G. (2013). Implementation of clinical practise guidelines for ventilator-
associated pneumonia: A multicenter prospective study. Critical Care Medicine, 41(1), 15-
23. doi:10.1097/CCM.0b013e318265e874
Smiddy, M. P., O' Connell, R., & Creedon, S. A. (2015). Systematic qualitative literature review of
health care workers' compliance with hand hygiene guidelines. American Journal of Infection
Control, 43(3), 269-274. doi:10.1016/j.ajic.2014.11.007
Smith, J., Swallow, V., & Coyne, I. (2015). Involving parents in managing their child & long-term
condition—A concept synthesis of Family-Centered Care and Partnership-in-Care. Journal of
Pediatric Nursing, 30(1), 143-159. doi:10.1016/j.pedn.2014.10.014
Spirt, M. J. (2003). Acid suppression in critically ill patients: What does the evidence support?
Pharmacotherapy: The Journal of Human Pharmacology and Drug Therapy, 23(10P2), 87S-
93S. doi:10.1592/phco.23.13.87S.31933
Srinivasan, R., Asselin, J., Gildengorin, G., Wiener-Kronish, J., & Flori, H. R. (2009). A prospective
study of ventilator-associated pneumonia in children. Pediatrics, 123(4), 1108-1115.
doi:10.1542/peds.2008-1211
Stabouli, S., Volakli, E., Violaki, A., Tsolaki, A., Krommyda, M., & Sdougka, M. (2012). 426
Ventilator-Associated Pneumonia in Children: A Retrospective Analysis. Archives of Disease
in Childhood, 97(Suppl 2), A125-A125. doi:10.1136/archdischild-2012-302724.0426
StataCorp. (2015). Stata Statistical Software: Release 14. College Station, TX: StataCorp L.P
Stein, D., Bassel, M., & Payne, K. A. (2014). Retrospective chart review studies: Strategies to ensure
robust data quality. Value Health, 17(7), A583. doi:10.1016/j.jval.2014.08.1979
Stevens, J. P., Silva, G., Gillis, J., Novack, V., Talmor, D., Klompas, M., & Howell, M. D. (2014).
Automated surveillance for ventilator-associated events. Chest, 146(6), 1612-1618.
doi:10.1378/chest.13-2255
Stoeppel, C. M., Eriksson, E. A., Hawkins, K., Eastman, A., Wolf, S., Minei, J., & Minshall, C. T.
(2014). Applicability of the National Healthcare Safety Network's surveillance definition of
ventilator-associated events in the surgical intensive care unit: A 1-year review. Journal of
Trauma Acute Care Surgery, 77(6), 934-937. doi:10.1097/TA.0000000000000425
Stickney, C. A., Ziniel, S. I., Brett, M. S., & Truog, R. D. (2014). Family Participation during
Intensive Care Unit Rounds: Attitudes and Experiences of Parents and Healthcare Providers
in a Tertiary Pediatric Intensive Care Unit. Journal of Pediatrics, 164(2), 402-402. Retrieved
from doi:10.1016/j.jpeds.2013.09.037 Straney, C. L., Clements, C. A., Parslow, C. R., Pearson, C. G., Shann, C. F., Alexander, C. J., &
Slater, C. A. (2013). Paediatric Index of Mortality 3: An Updated model for predicting
mortality in pediatric intensive care*. Pediatric Critical Care Medicine, 14(7), 673-681.
doi:10.1097/PCC.0b013e31829760cf
Tabaeian, S. M., Yazdannik, A., & Abbasi, S. (2017). Compliance with the standards for prevention
of ventilator-associated pneumonia by nurses in the intensive care units. Iranian Journal of
Nursing and Midwifery Research, 22(1), 31. doi:10.4103/1735-9066.202073
Tablan, O. C., Anderson, L. J., Besser, R., Bridges, C., & Hajjeh, R. (2004). Guidelines for preventing
healthcare-associated pneumonia, 2003. MMWR, 53(RR-3), 1-36.
Talbot, T. R., Johnson, J. G., Fergus, C., Domenico, J. H., Schaffner, W., Daniels, T. L., . . . Hickson,
G. B. (2013). Sustained improvement in hand hygiene adherence: Utilizing shared
accountability and financial incentives. Infection Control and Hospital Epidemiology, 34(11),
1129-1136. doi:10.1086/673445
Talekar, C. R., Udy, A. A., Boots, R. J., Lipman, J., & Cook, D. (2014). Tracheal cuff pressure
monitoring in the ICU: A literature review and survey of current practise in Queensland.
Anaesthesia and Intensive Care, 42(6), 761-770. doi: 10.1177/0310057x1404200612
Tamma, P. D., Turnbull, A. E., Milstone, A. M., Lehmann, C. U., Sydnor, E. R. M., & Cosgrove, S.
E. (2011). Ventilator-associated tracheitis in children: Does antibiotic duration
matter?(Clinical report). Clinical Infectious Diseases, 52(11), 1324. doi:10.1093/cid/cir203
186
Tang, C. W., Liu, P. Y., Huang, Y. F., Pan, J. Y., Lee, S., Hsieh, K. S., . . Ger, L. P. (2009). Ventilator-
associated pneumonia after pediatric cardiac surgery in southern Taiwan. Journal of
Microbiology, Immunology and Infection, 42(5), 413-419.
Taylor, C. N., Noronha, L., Wichman, C. S., & Varman, M. (2014). Evaluation of 2 sets of screening
criteria for ventilator-associated pneumonia in a children's hospital. American Journal of
Infection Control, 42(9), 1011-1013.
The Joint Commission. (2018). Speak up: Prevent the spread of infection. Retrieved from
https://www.jointcommission.org/multimedia/speak-up--prevent-the-spread-of-infection/
The National Health and Medical Research Council (NHMRC). (2000). NHMRC additional levels of
evidence and grades for recommendations for developers of guidelines. Retrieved from
https://www.nhmrc.gov.au/_files_nhmrc/file/guidelines/developers/nhmrc_levels_grades_ev
idence_120423.pdf
Tobias, J. D., Schwartz, L., Rice, J., Jatana, K., & Kang, D. R. (2012). Cuffed endotracheal tubes in
infants and children: should we routinely measure the cuff pressure? International Journal of
Pediatric Otorhinolaryngology, 76(1), 61-63. doi:10.1016/j.ijporl.2011.09.033
Tolentino-DelosReyes, A. F., Ruppert, S. D., & Shiao, S. Y. P. K. (2007). Evidence-based practise:
Use of the ventilator bundle to prevent ventilator-associated pneumonia. American Journal of
Critical Care, 16(1), 20-27. Retrieved from http://ajcc.aacnjournals.org/.
Topeli, A., Harmanci, A., Cetinkaya, Y., Akdeniz, S., & Unal, S. (2004). Comparison of the effect of
closed versus open endotracheal suction systems on the development of ventilator-associated
pneumonia. Journal of Hospital Infection, 58(1), 14-19. doi:10.1016/j.jhin.2004.05.005
Tume, L. N., & Copnell, B. (2015). Endotracheal suctioning of the critically ill child. Journal of
Pediatric Intensive Care, 4(02), 056-063. doi: 10.1055/s-0035-1556747
Turton, P. (2008). Ventilator-associated pneumonia in paediatric intensive care: A literature review.
Nursing in critical care, 13(5), 241-248. doi:10.1111/j.1478-5153.2008.00290.x
Uhl, T., Fisher, K., Docherty, S. L., & Brandon, D. H. (2013). Insights into Patient and Family‐Centered Care through the hospital experiences of parents. Journal of Obstetric, Gynecologic,
& Neonatal Nursing, 42(1), 121-131. doi:10.1111/1552-6909.12001
University Hospitals Bristol NHS Foundation Trust. (2009). How to: Set and Audit Sample & Plan
Your Data Collection. Retrieved from http://www.uhbristol.nhs.uk/for-
clinicians/clinicalaudit/how-to-guides/
Venkatachalam, V., Hendley, J. O., & Willson, D. F. (2011). The diagnostic dilemma of ventilator-
associated pneumonia in critically ill children. Pediatric Critical Care Medicine, 12(3), 286-
296. doi:10.1097/PCC.0b013e3181fe2ffb
Vet, N. J., de Wildt, S. N., Verlaat, C. W. M., Knibbe, C. A. J., Mooij, M. G., van Woensel, J. B. M.,
. . de Hoog, M. (2016). A randomized controlled trial of daily sedation interruption in critically
ill children. Intensive Care Medicine, 42(2), 233-244. doi:10.1007/s00134-015-4136-z
VICNISS Healthcare Associated Infection Surveillance. (2015). VICNISS Surveillance Activities in
our Hospitals. Retrieved from https://www.vicniss.org.au/about/surveillance-activities/
Volsko, T. A. (2013). Airway clearance therapy: Finding the evidence. Respiratory Care, 58(10),
1669-1678. doi:10.4187/respcare.02590
Voos, K. C., & Park, N. (2014). Implementing an open unit policy in a neonatal intensive care unit:
nurses' and parents' perceptions. Journal of Perinatal and Neonatal Nursing, 28(4), 313-318.
doi:10.1097/JPN.0000000000000055
Webb, P., Bain, C., & Page, A. (2017). Essential epidemiology: An introduction for students and
health professionals 3rd ed.). Cambridge: Cambridge University Press.
Weinger, M. B., Slagle, J., Jain, S., & Ordonez, N. (2003). Retrospective data collection and
analytical techniques for patient safety studies. Journal of Biomedical Informatics, 36(1-2),
106-119. doi:10.1016/j.jbi.2003.08.002
Weisberg, S. (2005). Applied linear regression. Hoboken. NJ: John Wiley & Sons.
Weiss, M., & Dullenkopf, A. (2007). Cuffed tracheal tubes in children: Past, present and future.
Expert Review of Medical Devices, 4(1), 73-82. doi:10.1586/17434440.4.1.73
187
Weiss, M., Dullenkopf, A., Fischer, J. E., Keller, C., Gerber, A. C., & European Paediatric
Endotracheal Intubation Study, G. (2009). Prospective randomized controlled multi-centre
trial of cuffed or uncuffed endotracheal tubes in small children. British Journal of
Anaesthesia, 103(6), 867-873. doi:10.1093/bja/aep290
Wip, C., & Napolitano, L. (2009). Bundles to prevent ventilator-associated pneumonia: How valuable
are they? Current Opinion in Infectious Diseases, 22(2), 159-166.
doi:10.1097/QCO.0b013e3283295e7b
Wong, A., Mathieu, S., & Williams, M. (2012). Is ventilator-associated pneumonia (VAP) an
appropriate quality indicator? Journal of the Intensive Care Society, 13(1), 79-81.
doi:10.1177/175114371201300126
Woodward, M. (2014). Epidemiology study design and data analysis (3rd ed.). Boca Raton: CRC
Press.
World Health Organization (WHO). (2009a). A guide to the implementation of the WHO multimodal
hand hygiene improvement strategy. Retrieved from https://www.who.int/.
World Health Organization (WHO). (2009b). WHO Guidelines on Hand Hygiene in Health Care.
Retrieved from https://ci.nii.ac.jp/naid/10030447988/en/.
World Health Organization (WHO). (2017). Clean Care is Safer Care: Tools for evaluation and
Feedback. Retrieved from http://www.who.int/gpsc/5may/tools/evaluation_feedback/en/
Wright, M. L., & Romano, M. J. (2006). Ventilator-associated pneumonia in children. Seminars in
Pediatric Infection Diseases, 17(2), 58-64. doi:10.1053/j.spid.2006.04.003
Wu, K. S., Lee, S. S., Chen, J. K., Tsai, H. C., Li, C. H., Chao, H. L., . . Chen, Y. S. (2013). Hand
hygiene among patients: Attitudes, perceptions, and willingness to participate. American
Journal of Infection Control, 41(4), 327-331. doi:10.1016/j.ajic.2012.03.033
Yildizdas, D., Yapicioglu, H., & Yilmaz, H. L. (2002). Occurrence of ventilator-associated
pneumonia in mechanically ventilated pediatric intensive care patients during stress ulcer
prophylaxis with sucralfate, ranitidine, and omeprazole. Journal of Critical Care, 17(4), 240-
245. doi:10.1053/jcrc.2002.36761
Yokoe, D. S., Anderson, D. J., Berenholtz, S. M., Calfee, D. P., Dubberke, E. R., Ellingson, K. D., .
. Klompas, M. (2014). A compendium of strategies to prevent healthcare-associated infections
in acute care hospitals: 2014 updates. American Journal of Infection Control, 42(8), 820-828.
doi: 10.1016/j.ajic.2014.07.002
Zolfaghari, P. S., & Wyncoll, D. L. A. (2011). The tracheal tube: gateway to ventilator-associated
pneumonia. Critical Care, 15(5), 310. doi:10.1186/cc10352
Zou, G. (2004). A modified poisson regression approach to prospective studies with binary data.
American Journal of Epidemiology, 159(7), 702-706. doi: 10.1093/aje/kwh090
188
Appendices
Appendix A Data collection sheet (Retrospective study)
No Variable Label Definition Coding in SPSS Remarks 1. Patient no Pt no Patient no labelled after met the
inclusion & exclusion criteria [Cohort
2015]
Patient 1 to 234
1
2
3 and until 234
Corresponding to the
number of patients in the
study
2. Episodes of
mechanical
ventilation
EpisodesofMV_exp
erimental_unit
The number is corresponding to the
number of mechanical episodes of
each patient
1- 262 The experimental unit
3. Admission Admission Number of PICU admission of each
patient
1 for 1st admission
2 for 2nd admission
3 for 3rd admission
Demographic
characteristics
4. Mechanical
ventilation
number
MVnumber Mechanical ventilation number
during admission for each patient
1 for MV number 1 (single)
2 for MV number 2
3 for MV number 3
4 for MV number 4
5. Age Age_in months Age of the patients - in months True age (in 2 decimal points) –
in months
Demographic
characteristics & /risk
factor 6. Gender Gender The sex of patients – male and female 1. Male
2. Female
Demographic
characteristics & /risk
factor 7. Weight Wt Weight of patients- in Kg True weight in Kg Demographic
characteristics & /risk
factor 8. PICU admission
source
IADM_SC_PICU The classification is based on
ANZPIC registry data collection
form- 2016
1. OT/recovery,
2. Emerg Dept
3. Ward (any other inpatient area)
4. Other ICU/NICU-same
hospital
5. Direct ICU Adm’n
Demographic
characteristics & /risk
factor
189
9. Underlying
disease for
PICU admission
UDX The codes are according to ANZPIC
Diagnostic Codes Table- 2015
1. Injury/trauma (100-
2. Cardiovascular (200-
3. Neurological (300-
4. Respiratory (400-
5. Renal (500-
6. Gastrointestinal (600-
7. Infection (700-
8. Miscellaneous (800-
9. Post- procedural (1100-
Demographic
characteristics/ risk
factor
*Requirement for CDC
PNEU/VAP surveillance
tool
10. PICU_diagnosis
_cat
PICU_diagnosis_ca
tegory
The codes are according to ANZPIC
Diagnostic Codes Table- 2015
1. Medical
2. Surgical
3. Trauma
Demographic
characteristics/ risk
factor
11. PIMS 2 score PIMS2_DEATH Paediatric index of mortality score True score Demographic
characteristics & /risk
factor 12. Paediatric index
of mortality 2
PIMS2_Anz11_DE
ATH
Calibrated Paediatric index of
mortality score_anz2011
True score Demographic
characteristics & /risk
factor 13. PIMS 3 score PIMS3_DEATH Paediatric index of mortality score True score Demographic
characteristics & /risk
factor 14. Paediatric
index_anz2013
of mortality 3
PIMS3_anz13_DE
ATH
Calibrated Paediatric index_anz2013
of mortality score
True score Demographic
characteristics & /risk
factor 15. Commencement
of mechanical
ventilation
Commenceof MV Date and time commencement of
mechanical ventilation retrieved from
ANZPIC via Metavision
Date and time in the format of
dd/mm/2017 hh:mm (24hrs
system)
Commencement of
mechanical ventilation
16. Cease of
mechanical
ventilation
CeaseofMV Date and time cease of mechanical
ventilation retrieved from ANZPIC
via Metavision
Date and time in the format of
dd/mm/2017 hh:mm (24hrs
system)
Cease of mechanical
ventilation
17. Column ceased
commenced 1st
Columnceasedcom
menced1stidentifed
xray
Date and time of 1st identified as
ceased of mechanical ventilation for
those case that did not identified as
Date and time in the format of
dd/mm/2017 hh:mm (24hrs
system)
Column ceased
commenced 1st identified
X-Ray_VAP
190
identified X-
Ray_VAP
VAP (same as date and time of ceased
of mechanical ventilation) 18. Time at risk of
VAP
TimeatriskofVAP Total of hours
Columnceasedcommenced1stidentife
dxray minus Commenceof MV
In total of hours (hh: mm: ss) Time at risk of VAP
19. Column ceased
commenced 1st
identified VAC
Columnceasedcom
menced1stidentifea
s VAC
Date and time of 1st identified as ceased of
mechanical ventilation for those case that did
not identified as VAC (same as date and time
of ceased of mechanical ventilation)
Date and time in the format of
dd/mm/2017 hh:mm (24hrs
system)
Column ceased
commenced 1st identified
VAC
20. Time at risk of
VAE
TimeatriskofVAE Total of hours
Columnceasedcommenced1stidentife
dVAC minus Commenceof MV
In total of hours (hh: mm: ss) Time at risk of VAE
21. Duration of MV Duration_MV Recorded in days (commence date &
time minus cease date & time of
intubation
True numbers of days Outcomes
22. Duration of
PICU stay
(days)
Length_of_PICU_s
tay
Recorded in days (PICU admission
date & time minus PICU discharge
date & time)
True numbers of days Outcomes
23. Duration of
hospital stay
(days)
Length_of_hospital
_stay
Recorded in days (Hospital admission
date & time minus hospital discharge
date& time)
True numbers of days Outcomes
24. PICU outcomes PICU_OUTCOME The classification are based on
ANZPIC registry data collection
form- 2016:
1- Discharge to ward/home;
2- Died in ICU;
3- Transferred to another ICU
(includes NICU),
4- Still in ICU,
5- Died within 24 hours after
being discharged from ICU to
receive palliative care
Outcomes
25. Mortality Mortality Died or not died 0- Not died
1- Died
26. Ventilator-
associated
VAP yes or no Number of identified as VAP case-
meeting the CDC tool from each
0- No
1- Yes
Manually determined
based on CDC
191
pneumonia
(VAP)
experimental unit (from 261 episodes
of MV)
PNEU/VAP surveillance
tool 27. Ventilator-
associated event
(VAE)
VAE classifications Number of identified as VAE (VAC,
IVAC, PoVAP - meeting the VAE
algorithm
0=No
1. VAC
2. IVAC
3. PoVAP
Manually calculated/
determined using new
CDC VAE surveillance
tool
CDC VAE calculator
version 4.0 then used to
confirmation. 28. Mechanical
ventilation days
MV_day MV days that each patient was on
ETT tube. Each patient may have
varied number of MV days
Data from Day 1…max day 49
(in this cohort) documented)
29. Reintubation
episodes
Reintubationepisod
es
The presence of any reintubation was
taken place or not.
0- No
1- Yes
Risk factor_ raw data
30. Overall
Reintubation
episodes_for
risk factor
analysis
Overall_Reintubati
onepisodes
If any event of reintubation took place
in each days of every episodes (1)
consider as yes and (0) is no.
0- No
1- Yes
Risk factor – will use to
further analysis
31. Routes of
intubation
Routes_Intubation Routes of endotracheal intubation
whether nasal or oral
1- Nasal
2- Oral
Risk factor_ raw data
32. Overall routes
of intubation
Overall_Routes_Int
ubation
Determined as with Nasal tube if the
patient has the longest (MV days) on
nasal tube.=coded as 1
Determined as with Oral tube if
patient has the longest (MV days) on
Oral tube=coded as 2
If the same total numbers of MV days
– consider the last changed/ the
newest type of tube to be overall
1- Nasal
2- Oral
Risk factor – will use to
further analysis
33. Highest sedation
scores
Highestsedationsco
re
-3 – unresponsive/paralysed
-2 – response to noxious stimuli
-1 – response to gentle touch/voice
0 – awake & able to calm
-3
-2
-1
0
Risk factor_ raw data
192
+1 – restless & difficult to calm
+2- agitated
+1
+2 34. Overall Highest
sedation score
OverallHighestseda
tionscore
The score is counted as overall if the
score had the majority numbers of
MV days in episode/s of MV. If the
score found in equal number of MV
days; the score on the last day will be
taken as overall
Recoded into 3 categories
1- Deep Sedation
(Score of -2 to -3)
2 – Light Sedation
(Score of -1 to +1)
3- Agitated
(Score of +2)
Risk factor – will use to
further analysis
35. Paralytic agent paralyticagent The administered of any paralytic
drugs for each MV days
yes - if the infusions were given> 4
hours of infusion or > 4 times bolus
dose
no – other than the criteria above
0- No
1- Yes
Risk factor_ raw data
36. Overall
Paralytic agent
Overall_Paralyticag
ent
Overall Yes (1) is determined if the
“yes” found in the majority of MV
days of every episodes of MV or if the
“yes” are equal with “no” in MV days
– counted as Yes (1)
0- No
1- Yes
Risk factor – will use to
further analysis
37. Gastrointestinal
prophylaxis
GIprophylaxis The administered of any GI
prophylaxis drugs for each MV days
0- No
1- Yes
Risk factor_ raw data
38. Overall GI
prophylaxis
OverallGiprophylax
is
Overall Yes (1) is determined if the
“yes” found in the majority of MV
days of every episodes of MV or if the
“yes” are equal with “no” in MV days
– counted as Yes (1)
0- No
1- Yes
Risk factor – will use to
further analysis
39. Antibiotic What_antibiotic Name of antibiotics, which the
patients were administered for each
MV days
The antibiotics are based on list stated
in VAE surveillance tool (67
antibiotics) refer to antibiotic list at
the end of sheet).
Recorded as name of respective
antibiotic/s
NA – if the patients were on
antibiotics that not in the list
No_Ab- if the patients were not
on any antibiotic/s.
Requirement for VAE;
IVAC (2nd Tier) [new
CDC VAE surveillance
tool]
193
40. Number of
antibiotics
Number_of_antibio
tic
Number of antibiotics corresponding
to list of antibiotics stated in VAE
surveillance tool
1- one antibiotic
2 – 2 antibiotics
3- 3 antibiotics
4- 4 antibiotics
5- 5 antibiotics
6- NA
7- No antibiotic
41. Nasogastric
(NG)tube
presence
Nasogastrictubepre
sence
The presence of NG tube for each MV
days
0- No
1- Yes
Risk factor_ raw data
42. Overall
Nasogastric tube
presence
OverallNasogastrict
ubepresence
Overall Yes (1) is determined if the
“yes” found in the majority of MV
days of every episodes of MV or if the
“yes” are equal with “no” in MV days
– counted as Yes (1)
0- No
1- Yes
Risk factor – will use to
further analysis
43. X-Rays with
disease
X-Ray with disease Yes - if radiological report (in each
MV days) distinguished the findings
as: 2 or more serial X-Rays with 1 of
the following ✓ new or progressive and
persistent infiltrate
✓ consolidation
✓ cavitation
✓ pneumatoceles, in < 1 y.o
No – if none of radiological report (in
each MV days) found as the list of X-
Rays findings and no X-Ray done on
the particular days
NA- if patient without disease
0- No
1- Yes
3- NA
Requirement for CDC
PNEU/VAP surveillance
tool
44. X-Rays without
disease
X-Ray without
disease
Yes - if radiological report (in each
MV days) distinguished the findings
as: 1 or more serial X-Rays with 1 of
the following
0- No
1- Yes
3- NA
Requirement for CDC
PNEU/VAP surveillance
tool
194
✓ new or progressive and
persistent infiltrate
✓ consolidation
✓ cavitation
✓ pneumatoceles, in < 1 y.o
No – if none of radiological report (in
each MV days) found as the list of X-
Rays findings and no X-Ray done on
the particular days
NA- if patient with disease 45. Worsening gas
exchange
[(Increased
oxygen
requirement, or
increased
ventilation
demand)]
(i)FiO2 (ii)
PEEP or iii)
Spo2 <94%
Min_FiO2 True value of minimum FiO2 at least
sustained for 1 hour in a day -for
every MV days
True value of minimum FiO2 Requirement for VAC
(VAE) -1st Tier [new
CDC VAE surveillance
tool]
Additional parameter to
assess Worsening gas
exchange [(Increased
oxygen requirement, or
increased ventilation
demand in CDC
PNEU/VAP surveillance
tool 46. Worsening gas
exchange
[(Increased
oxygen
requirement, or
increased
ventilation
demand)]
(i)FiO2 (ii)
PEEP or iii)
Spo2 <94%
Min PEEP True value of minimum PEEP at least
sustained for 1 hour in a day -for
every MV days
True value of minimum PEEP,
9999 as missing value if no PEEP
value documented.
Requirement for VAC
(VAE)- 1st Tier [new
CDC VAE surveillance
tool]
Additional parameter to
assess Worsening gas
exchange [(Increased
oxygen requirement, or
increased ventilation
demand in CDC
PNEU/VAP surveillance
tool
195
47. Worsening gas
exchange
[(Increased
oxygen
requirement, or
increased
ventilation
demand)]
(i)FiO2 (ii)
PEEP or iii)
Spo2 <94%
SpO2 < 94% Yes- if any of SpO2 readings <94% in
a day- for every MV days
No
0- No
1- Yes
2- No value
Requirement for CDC
PNEU/VAP surveillance
tool especially for infant
<1y.o apart from
worsening oxygenation
SpO2 <94%
Also, for children >1y.o
48. Temperature
instability
Temp<36oC_or>38oC
Yes- if any of temperatures was
36oC_or>38oC in a day – for every
MV days.
No
0- No
1- Yes
2- No value
Requirement for both
CDC PNEU/VAP
surveillance tool and
VAE (IVAC- 2nd tier)
[new CDC VAE
surveillance tool] 49. Tachypnea
RR_aged_based Defined as [(> 75 breath/min –
premature infants born at <37 wks &
until 40th wks; >60bpm=<2 months
old; >50bpm= 2-12 months old;
>30bpm=children >1 yr. old)].
Yes – if any of the respiration rates in
a day of each MV days met the tool
based on age classifications as above.
No
0- No
1- Yes
2- No value
Requirement for CDC
PNEU/VAP surveillance
tool
50. Bradycardia or
tachycardia
HR<100or>170_inf
ant<1yr
Defined as HR <100 or
>170beats/min
Yes- if any of HR reading was <100
or >170beats/min in a day – for every
MV days.
No
NA (for infant >1.yr.o)
0- No
1- Yes
2- No value
3- NA for patients >1 yr.o.
Requirement for CDC
PNEU/VAP surveillance
tool
Applicable for infant
<1y.o only
196
51. Changes in
character of the
sputum
changesputumchara
c
Defined as change in sputum
character [(i) colour; (ii) consistency;
(iii) quantity
Yes
No
Considering this character first:
Sputum colour: yellow, creamy,
green, brown
Sputum consistency: thick
Sputum quantity: large, copious,
moderate
0- No
1- Yes
Requirement for CDC
PNEU/VAP surveillance
tool
52. Leukopenia or
Leukocytosis
WBCC (≤4K or
≥15K)
Yes- if any of WBCC results was ≤4K
or ≥15K in a day – for every MV days.
No
0- No
1- Yes
2- No value/no sample sent
Requirement for both
CDC PNEU/VAP
surveillance tool and
VAE (IVAC- 2nd-Tier)
[new CDC VAE
surveillance tool]
After gathered all data &
the VAC confirmed;
second screening done
for IVAC determination
as ≥12K is required 53. Microbiological
consideration of
respiratory
secretions
Gram_staning Recorded for 1. ETA
2. BAL
3. Pleural fluid
4. Lung tissue biopsy
Gram+ve, Gram-ve, no growth, no
sample sent :(cocci, bacilli or other)
(washing BAL or BAL) (Pleural fluid
RT/LT) (Lung tissue biopsy)
Recorded as in the microbiology
findings (same as in tool column)
Requirement for both
CDC PNEU/VAP
surveillance tool
and VAE (possible VAP
(Tier 3) [new CDC VAE
surveillance tool]
54. Corresponding
values to
quantitative
Colony_forming_u
nit
Documented as
Epi – scant, 1+ -4+
Leu- scant, 1+ - 4+
Recorded as in the microbiology
findings (same as in tool column)
Requirement for both
CDC PNEU/VAP
surveillance tool
197
threshold values
for cultured
specimens used
in diagnosis of
pneumonia
Sq. epi
Erythrocyte
WBC’s ___x10^6/L;
RBC’s___x10^6/L;
Polymorphs___%
Contains clots, blood stained
and VAE (possible VAP
(Tier 3) [new CDC VAE
surveillance tool]
55. Name of the
organism with
quantitative
threshold values
for cultured
specimens used
in diagnosis of
pneumonia
Organisms isolated No organism seen, normal respiratory
flora, organism name (scant, 1+ -4+),
no fungi isolate, coagulase neg.
staphylococcus
Isolate_____CFU/ML
Recorded as in the microbiology
findings (same as in tool column)
Requirement for both
CDC PNEU/VAP
surveillance tool
and VAE (possible VAP
(Tier 3) [new CDC VAE
surveillance tool]
56. HH_compliance Hand hygiene
compliance
Hand hygiene compliance in % for
each month in 2015 obtained from
Hand Hygiene Australia –
True % of hand hygiene, the
value with the respective months
that the patients were on
mechanical ventilation
Missing value 9999
Requirement for VAP
preventative strategies
57. Oral hygiene
practises
MouthCare_perfor
mance
Mouth care is referred to whether it
was performed or not during MV days
Yes- if documented any of oral
hygiene was performed at daily basis
regardless of mouth care plans.
No – is not done
0- No
1- Yes
Requirement for
preventive strategies
performed by nurses
58. Overall Mount
Care
Performance
Overall_MC_perfor
mance
Counted as the majority of Yes/ No in
each MV days. If equal number of
Yes and No found in one episode of
MV – considered as Yes
0- No
1- Yes
Represent whether the
preventive strategies
were performed or not
during each episode of
MV 59. Frequency of
oral hygiene
practises
MouthCareF Is referred to the frequency of mouth
care was performed for every MV
days
Frequency of MC performance
(count)
Raw data
198
60. Overall MC
Frequency
OverallMCFrequen
cy
Average of the frequency to MV days
of every episodes of MV
1 average value recorded
represent each episodesof MV
Further analysis required
to determine the
compliance rate based
PICU Oral Hygiene
protocol (<6 months
without teeth and > 6
months with teeth) 61. Endotracheal
(ET) suctioning
Suctioning_perform
ance
Suctioning via ET is referred to
whether it was performed or not
during MV days
0- No
1- Yes
Requirement for
preventive strategies
performed by nurses 62. Overall
Suctioning
performance
OverallSuctioning_
performance
Counted as the majority of Yes/ No in
each MV days. If equal number of
Yes and No found in one episode of
MV – considered as Yes
0- No
1- Yes
Represent whether the
preventive strategies
were performed or not
during each episode of
MV 63. Frequency of
ET suctioning
practises
SuctioningF Is referred to the frequency of ET
suction was performed for every MV
days
Frequency of ET performance
(count)
Raw data
64. Overall
suctioning
frequency
Overall_suctioning
_frequency
Average of the frequency to MV days
of every episodes of MV
1 average value recorded
represent each episode of MV
Further analysis required
to determine the
compliance rate based on
PICU minimum standard
of suctioning frequency
(4 times/ 24 hours) 65. Cuff pressure
checking (ETT
with cuff &
uncuffed)
Cuffpressurechecks
_performance
Cuff pressure check is referred to
whether it was performed or not
during MV days including patients on
uncuffed ET
0- No
1- Yes
3 – NA (no cuff or ordered as
deflated)
Requirement for
preventive strategies
performed by nurses
66. Overall Cuff
Pressure check
performance
OverallCuffPressur
echeck_performanc
e
Counted as the majority of Yes/ No/
NA in each MV days. If equal number
of Yes/ No /NA found in one episode
of MV – considered as Yes
0- No
1- Yes
3 – NA (no cuff or ordered as
deflated)
Represent whether the
preventive strategies
were performed or not or
NA for patients who were
on uncuff ETT/ ordered
199
as deflated during each
episode of MV 67. Frequency of
cuff pressure
checks practises
Cuffpressurechecks
F
Is referred to the frequency of cuff
pressure check was performed for
every MV days
Frequency of Cuff Pressure
performance (count)
Raw data
68. Overall cuff
pressure
frequency
Overallcuffpressure
_frequency
Average of the frequency to MV days
of every episodes of MV
1 average value recorded
represent each episode of MV
Further analysis required
to determine the
compliance rate based on
PICU standard (at least
12 hourly or twice daily) 69. Head of bed
elevation
performance
Headofbedelevation
_performance
Head of bed elevation is referred to
whether it was performed or not
during MV days including 0 degree
0- No
1- Yes
Requirement for
preventive strategies
performed by nurses
70. Overall HOB
performance
OverallHOB_perfor
mance
Counted as the majority of Yes/ No in
each MV days. If equal number of
Yes and No found in one episode of
MV – considered as Yes
0- No
1- Yes
Represent whether the
preventive strategies
were performed or not
during each episode of
MV 71. Frequency of
head of bed
elevation
practises
HeadofelevationF Is referred to the frequency of head of
bed elevation care was performed in a
day for every MV days
Frequency of HOB performance
(count)
Raw data
72. Overall HOB
frequency
Overall_HOB_freq
uency
Average of the frequency to MV days
of every episodes of MV
1 average value recorded
represent each episode of MV
Further analysis required
to determine the
compliance rate based on
PICU standard
200
Appendix B HREC approval
201
202
203
Appendix C Governance approval
204
Appendix D PHA approval
205
Appendix E University of Queensland Ethical approval
206
Appendix F Updated VAP education package (PICU staff)
1) Ventilator-associated pneumonia: Nursing interventions with VAP preventative strategies
2) VAP preventative Strategies (‘bundle’) poster
207
Appendix G Education for parents: Pamphlet
VAP: How I Can Help My Child in PICU?
What is VAP?
Although healthcare providers try to give the best possible care to reduce the risk of
infection, sometimes patients can develop an infection while in the hospital.
Any child who is on a ventilator (breathing machine) is at risk of developing pneumonia
(lung infection), sometimes called Ventilator-associated Pneumonia (VAP).
How does VAP happen?
When your child is on a ventilator, she/he may not be able to cough and remove their
saliva easily. Their immune system is not as strong when on a ventilator.
What can I do as a parent?
The two ways to help stop VAP are:
1. Hand Hygiene
2. Mouth Care
Hand Hygiene:
The best way to prevent infections is to clean your hands. It is as simple as:
✓ Rubbing your hands with hand gel found throughout the PICU. Rub for 20-30
seconds, if hands are not visibly soiled, and making sure you get in between fingers
and under jewellery
OR
✓ Washing your hands with soap and water for 40-60 seconds when hands are
visibly dirty or after using the toilet.
When do I clean my hands?
1. Before entering your child’s room.
2. Before touching your child.
3. After touching your child.
4. Before leaving your child’s room.
5. If you cough/ sneeze or touch your face.
Your contribution to care is valued
208
LCCH PICU encourages you to check if I washed my hands
and check if healthcare providers or visitors have washed their
hands when caring for your child.
Mouth Care
✓ You may be able to help to perform mouth care for your child while he/she is on
the ventilator.
✓ Please let your nurse know if you would like to be involved.
✓ Your child’s nurse will teach you how to do this safely.
✓ The breathing tube must be safely secured at all times.
✓ STOP and check with your nurse before cleaning the mouth.
✓ Make sure you perform Hand Hygiene before and after performing Mouth Cares.
How can I find out more about VAP?
Please speak to the nurses or doctors caring for your child in PICU.
209
Appendix H Survey Questionnaire (Parents)
We would like you to complete this survey. We hope to learn what you understand about chest
infections that can occur when your child is on a ventilator. We would also like to learn your opinion
about ‘speaking up’ for hand hygiene.
It should take you about 10 minutes to complete this questionnaire. Please read the questions carefully
and then respond spontaneously.
✓ There is no right or wrong on your answers.
✓ Your answers are anonymous and will be kept confidential.
✓ This questionnaire has three sections.
Section A: Demographic information
Please tick (√) on designated box.
1. What is your age?
Less than 30 years old
31- 40 year old
41- 50 years old
More than 51 year old
2. What is your sex?
Male
Female
3. What is your highest completed education?
Postgraduate Degree
Bachelor Degree
Certificate (I & II and III & IV)
High School
Didn’t complete High School
*Hand hygiene is the process of cleaning your hands. There are two methods of hand
hygiene: washing with soap and water or the use of hand gel/alcohol-based rub/sanitizer.
** Other healthcare workers in this study is refer to other than nurses; example is doctor,
physiotherapist, dietitian, occupational therapist and etc.
210
4. Are you employed/ working in the healthcare field?
Yes
No
5. Has your child been admitted in Paediatric Intensive Care (PICU) before this admission?
Yes
No
Unsure
6. Has your child had previous experience receiving breathing support using a ventilator in PICU
before this admission?
7.
Yes
No
Unsure
Section B: A pamphlet of Ventilator- Associated Pneumonia (VAP): How I Can Help My Child
in PICU?
This section relates to the information provided in the pamphlet and your perception on the
importance of hand hygiene.
[Pamphlet-hard copy given]
Please tick (√) on designated box.
8. Have you ever heard about ventilator- associated pneumonia (VAP) or chest infection related to
breathing support before?
Yes
No
Unsure
9. What method do you routinely use for your hand hygiene in PICU?
Hand gel
Hand wash
Combination
211
10. Please answer the following questions or statements by ticking (√) on the most appropriate
response.
Questions Strongly
Disagree
Disagree Neither
Agree
nor
Disagree
Agree Strongly
Agree
Do you feel the pamphlet was easy to
understand? O O O O O
Did the information in the pamphlet
make you concerned? O O O O O
Did you find that the information
encourages you to participate in your
child’s care?
O O O O O
Do you think that parents’ hand
hygiene is important in the prevention
of infection in hospital including
ventilator- associated pneumonia
(VAP) in PICU?
O O O O O
Do you feel that nurses’ hand hygiene
is important? O O O O O
Do you feel that other healthcare
workers’ hand hygiene is important? O O O O O
Do you feel that nurses in PICU wash
their hands enough? O O O O O
Do you feel that other healthcare
workers’ in PICU wash their hands
enough?
O O O O O
Section C: ‘Speaking up’ for hand hygiene
This section is related to your opinion regarding ‘Speaking up’ for hand hygiene
11. Please read the following statements and tick (√) to the most appropriate response.
Statements
Strongly
Disagree Disagree
Neither
Agree nor
Disagree Agree
Strongly
Agree
I would remind nurses to perform hand
hygiene if necessary. O O O O O
I would remind other healthcare
workers to perform hand hygiene if
they did not.
O O O O O
I am willing to be reminded by nurses
to perform hand hygiene if I did not. O O O O O
I am willing to be reminded by other
healthcare workers to perform hand
hygiene if I did not.
O O O O O
212
I think by speaking up for hand
hygiene, it would increase hand
hygiene practise among nurses.
O O O O O
I think by speaking up for hand
hygiene, it would increase hand
hygiene practise among other
healthcare workers.
O O O O O
12. What would stop you from reminding nurses to perform hand hygiene if you see they are not?
You may tick (√) more than one from the responses provided and/or type your response in a
space provided.
I would be too embarrassed.
I wouldn’t want it to affect the care of my child.
I wouldn’t want to interrupt them.
I don’t feel it is my place to question nurses.
Or other reasons; please state:
………………………………………………………………………………………………………
13. What would stop you from reminding other healthcare workers to perform hand hygiene if you
see they are not?
You may tick (√) more than one from the responses provided and/or type your response in a
space provided.
I would be too embarrassed.
I wouldn’t want it to affect the care of my child.
I wouldn’t want to interrupt them.
I don’t feel it is my place to question other healthcare workers.
Or other reasons; please state:
………………………………………………………………………………………………………
13. Your suggestions or comments to improve hand hygiene pratice in the unit?
…………………………………………………………………………………………………………
…………………………………………………………………………………………………………
---------------------------------------------------End of the questions--------------------------------------------
--
Thank you for completing this survey.
213
Appendix I Survey participant information sheets (Parents and Nurses)
214
215
216
217
Appendix J Survey Questionnaire (Nurses)
We would like you to complete this survey. We hope to learn what you understand about ventilator-
associated pneumonia (VAP) while taking care of your patients. We also hope to learn your opinion
about ‘speaking up’ for hand hygiene.
It should take you less than 5 minutes to complete this questionnaire. Please read the
questions carefully and then respond spontaneously.
✓ There is no right or wrong on your answers
✓ Your answers are anonymous and will be kept confidential
✓ This questionnaire has two sections
Section A: Demographic information
Please tick (√) on designated box and/ or write your response in a space provided.
1. What is your sex?
2. How long have you been working in PICU? (in months/years)
___________________________
3. What method do you routinely use for your hand hygiene in PICU?
Male
Female
Hand gel
Hand wash
Combination
*Hand hygiene is the process of cleaning your hands. There are two methods of hand
hygiene: washing with soap and water or the use of hand gel/alcohol-based rub/sanitizer.
** Other healthcare workers in this study is refer to other than nurses; example is doctor,
physiotherapist, dietitian, occupational therapist and etc.
218
Section B: ‘Speaking up’ for hand hygiene
This section is related to your opinion regarding 'Speaking up' for hand hygiene.
4. Please read the following statements and tick (√) to the most appropriate response.
Statements
Strongl
y
Disagre
e
Disagre
e
Neither
Agree nor
Disagree
Agre
e
Strongly
Agree
I would remind parents to perform hand
hygiene if necessary. O O O O O
I am willing to be reminded by parents to
perform hand hygiene if I did not. O O O O O
I am willing to be reminded by other
healthcare workers to perform hand
hygiene if I did not.
O O O O O
I think by speaking up for hand hygiene,
it would increase hand hygiene practise
among parents.
O O O O O
I think by speaking up for hand hygiene,
it would increase hand hygiene practise
among other healthcare workers.
O O O O O
5. What would stop you from reminding parents to perform hand hygiene if you see they are not?
You may tick (√) more than one from the responses provided and/or type your response in a
space provided.
I would be too embarrassed.
I wouldn’t want to interrupt them.
I don’t feel it is my place to question parents.
Or other reasons; please state:
_________________________________________________________________________
6. What would stop you from reminding other healthcare workers to perform hand hygiene if you
see they are not?
You may tick (√) more than one from the responses provided and/or type your response in a
space provided.
I would be too embarrassed.
I wouldn’t want to interrupt them.
I don’t feel it is my place to question them.
Or other reasons; please state:
_________________________________________________________________________
7. Your suggestions or comments to improve hand hygiene pratice in the unit.
________________________________________________________________________
________________________________________________________________________
Thank you for completing this survey.
219
Appendix K Data collection sheet (Prospective study)
No Variable Label Tool Coding in SPSS Remarks
1. Patient no Pt no Patient no labelled after met the inclusion
& exclusion criteria [Cohort 2017- six
months 12_6_17 till 12_12_17)]
Patient 1 to ….110
1
2
3 and until ….110
Corresponding to the
number of patients in the
study
2. Episodes of
mechanical
ventilation
EpisodesofMV_experi
mental_unit
The number is corresponding to the
number of mechanical episodes of each
patient
1- …120 The experimental unit
_episodes of MV evaluated
for VAP and VAE
3. Admission PICU_Admission Number of PICU admission of each
patient
1 for 1st admission
2 for 2nd admission
3 for 3rd admission
Demographic
characteristics
4. Mechanical
ventilation number
MVnumber Mechanical ventilation number during
admission for each patient
1 for MV number 1 (single)
2 for MV number 2
3 for MV number 3
5. Patient’s admission MV_Pt_admission
Number of MV in each admission
1 for MV no 1 in admission no 1
2 for MV no 2 in admission no 1
….
6. Age Age_in months Age of the patients - in months True age (in 1 decimal point) –
in months
Demographic
characteristics & /risk
factor
7. Gender Gender The sex of patients – male and female Male
Female
Demographic
characteristics & /risk
factor
220
8. Weight Wt Weight of patients- in Kg True weight in Kg (1 decimal
point)
Demographic
characteristics & /risk
factor
9. PICU admission
source
IADM_SC_PICU The classification is based on ANZPIC
registry data collection form- 2016
1. OT/recovery,
2. Emerg Dept
3. Ward (any other inpatient
area)
4. Other ICU/NICU-same
hospital
5. Direct ICU Adm’n
Demographic
characteristics & /risk
factor
10. IADM_SC_PICU_r
ecats_to4
Recats to 4 categories Re-categorised PICU to 4 categories only 1. OT/recovery,
2. Emerg Dept
3. Ward (any other inpatient
area); Other ICU/NICU-same
hospital
4. Direct ICU Adm’n
11. Principal ICU
diagnosis
PDX Code the diagnosis most directly
responsible for the ICU admission
1. Injury/trauma (100-
2. Cardiovascular (200-
3. Neurological (300-
4. Respiratory (400-
5. Renal (500-
6. Gastrointestinal (600-
7. Infection (700-
8. Miscellaneous (800-
9. Post- procedural (1100-
12. Underlying disease
for PICU admission
UDX The codes are according to ANZPIC
Diagnostic Codes Table- 2017
1. Injury/trauma (100-
2. Cardiovascular (200-
3. Neurological (300-
4. Respiratory (400-
5. Renal (500-
6. Gastrointestinal (600-
7. Infection (700-
8. Miscellaneous (800-
Demographic
characteristics/ risk factor
*Requirement for CDC
PNEU/VAP surveillance
tool
221
9. Post- procedural (1100-
13. UDX_recat_3cats Recategories UDX to 3
main cats
The codes are according to ANZPIC
Diagnostic Codes Table- 2017
1. cardiovascular
2. respiratory
3. others
14. PIMS 3 Paediatric Index of
Mortality
True calibrated PIMS 3 score True scores – 999- missing
value
15. Immunosuppressive
patient?
Immunosuppressive
pt?
Whether the patient is immunosuppressive
or not; eg: cancer patients, transplant
patients
Yes- 1
No- 0
Demographic
characteristics
16.
PICU Diagnosis
category
PICU
Diagnosis_category
Medical, surgical and Trauma 1- Medical
2 - Surgical
3- Trauma
Demographic characteristic
&/ risk factor
17. Diagnosis written Dignosis_written Diagnosis written in the daily report
available at CHQ server
Diagnosis written in the
Metavision or in daily PICU
census
Demographic characteristic
18. PICU date and time
of admission
PICU_admission PICU date and time of admission
retrieved from ANZPIC via Metavision
Date and time in the format of
dd/mm/2017 hh:mm (24hrs
system)
19. PICU date and time
discharged
PICU_Discharged PICU date and time of discharged
retrieved from ANZPIC via Metavision
Date and time in the format of
dd/mm/2017 hh:mm (24hrs
system)
20. Commencement of
mechanical
ventilation
Commenceof MV Date and time commencement of
mechanical ventilation retrieved from
ANZPIC via Metavision
Date and time in the format of
dd/mm/2017 hh:mm (24hrs
system)
21. Cease of
mechanical
ventilation
CeaseofMV Date and time cease of mechanical
ventilation retrieved from ANZPIC via
Metavision
Date and time in the format of
dd/mm/2017 hh:mm (24hrs
system)
22. Column ceased
commenced 1st
identified X-
Ray_VAP
Columnceasedcommen
ced1stidentifedxray
Date and time of 1st identified as ceased of
mechanical ventilation for those case that
did not identified as VAP (same as date
and time of ceased of mechanical
ventilation)
Date and time in the format of
dd/mm/2017 hh:mm (24hrs
system)
222
23. Column ceased
commenced 1st
identified VAC
Columnceasedcommen
ced1stidentifeas VAC
Date and time of 1st identified as ceased of
mechanical ventilation for those case that
did not identified as VAC (same as date
and time of ceased of mechanical
ventilation)
Date and time in the format of
dd/mm/2017 hh:mm (24hrs
system)
24. Duration of MV Duration_MV Recorded in days (MV cease date & time
minus MV commence date & time)
True numbers of days Outcomes
25. Duration of PICU
stay (days)
Length_of_PICU_stay Recorded in days (PICU discharge date &
time minus PICU admission date minus &
time)
True numbers of days Outcomes
26. Admission in
hospital
Admission_Hosp Date and time admission to hospital
retrieved from ANZPIC via Metavision
Date and time in the format of
dd/mm/2017 hh:mm (24hrs
system)
27. Discharged from
hospital
Discharged_Hosp Date and time discharged from hospital
retrieved from ANZPIC via Metavision
Date and time in the format of
dd/mm/2017 hh:mm (24hrs
system)
28. Duration of hospital
stay (days)
Length_of_hospital_sta
y
Recorded in days (Hospital discharge date
& time minus hospital admission date &
time)
True numbers of days Outcomes
29. PICU outcomes PICU_OUTCOME The classification is based on ANZPIC
registry data collection form- 2016:
1- Discharge to ward/home;
2- Died in PICU;
3- Transferred to another ICU
(includes NICU), or other
hospital
4- Still in PICU,
5- Died within 24 hours after
being discharged from ICU to
receive palliative care
Outcomes
30. Mortality Mortality_died and not
died
Categorised whether pt died or not 0- not died
1-died
31. PICU_outcome_rec
at_3cats
Recategorised to 3
categories
Recategorised to Discharge to ward/home;
Transferred to another ICU (includes
NICU), or other hospital, died in PICU
and died
1. Discharge to ward/home;
Transferred to another ICU
(includes NICU), or other
hospital; Died within 24 hours
223
after being discharged from ICU
to receive palliative care
2. Died in ICU
3. Still in PICU
32. Mechanical
ventilation days
MV_day MV days that each patient was on ETT
tube. Each patient may have varied
number of MV days
Data from Day 1…max day 61
(in this cohort) documented
33. Date of intubation Date Day 1 of patient on ETT Recorded as date eg: 12/06/2017 Important for VAE (VAC,
IVAC, PVAP is
determined)
34. Worsening gas
exchange
[(Increased oxygen
requirement, or
increased
ventilation
demand)] (i)FiO2
(ii) PEEP or iii)
Spo2 <94%
FiO2 _min True value of minimum FiO2 at least
sustained for 1 hour in a day -for every
MV days
True value of minimum FiO2,
999 as missing value if no FiO2
documented.
Requirement for VAC
(VAE) -1st Tier [new CDC
VAE surveillance tool]
Additional parameter to
assess Worsening gas
exchange [(Increased
oxygen requirement, or
increased ventilation
demand in CDC
PNEU/VAP surveillance
tool
35. Worsening gas
exchange
[(Increased oxygen
requirement, or
increased
ventilation
demand)] (i)FiO2
(ii) PEEP or iii)
Spo2 <94%
Min PEEP True value of minimum PEEP at least
sustained for 1 hour in a day -for every
MV days
True value of minimum PEEP,
999 as missing value if no PEEP
value documented.
Requirement for VAC
(VAE)- 1st Tier [new CDC
VAE surveillance tool]
Additional parameter to
assess Worsening gas
exchange [(Increased
oxygen requirement, or
increased ventilation
demand in CDC
PNEU/VAP surveillance
tool
224
36. Minimum
temperature
Temp_min True value of min temperature of the day True value of minimum temp,
999 as missing value if no mi
temp value documented.
IVAC and VAP
determination
37. Maximum
temperature
Temp_max True value of max temperature of the day True value of maximum PEEP,
999 as missing value if no max
temp value documented.
IVAC and VAP
determination
38. Minimum WBC WBC_min True value of min WBC of the day True value of minimum WBC,
999 as missing value if no min
WBC value documented.
IVAC and VAP
determination
39. Maximum WBC WBC_min True value of max WBC of the day True value of maximum WBC,
999 as missing value if no max
WBC value documented.
IVAC and VAP
determination
40. Antibiotic What_antibiotic Name of antibiotics, which the patients
were administered for each MV days
The antibiotics are based on list stated in
VAE surveillance tool (67 antibiotics) -refer to antibiotic list at the end of sheet)
Recorded as name of respective
antibiotic/s
NA – if the patients were on antibiotics
that not in the list
No_Ab- if the patients were not on any
antibiotic/s.
Requirement for VAE;
IVAC (2nd Tier) [new CDC
VAE surveillance tool]
41. Respiratory
specimens
Specimen Document the respiratory secretions sent
ETA
BAL
Pleural fluid
Lung tissue biopsy
Same as in the tool column PVAP determination and
supporting for VAP
(microbiological findings)
42. Polys/epis polys/Epis Documented the gram stain
Gram+ve, Gram-ve, no growth, no sample
sent: (cocci, bacilli or other) (washing
BAL or BAL) (Pleural fluid RT/LT)
(Lung tissue biopsy)
Documented as
Epi – scant, 1+ -4+
Leu- scant, 1+ - 4+
Same as in the tool column PVAP determination and
supporting for VAP
(microbiological findings)
225
Sq. epi
Erythrocyte
WBC’s ___x10^6/L; RBC’s___x10^6/L;
Polymorphs___%
Contains clots, blood stained
No organism seen, normal respiratory
flora, organism name (scant, 1+ -4+), no
fungi isolate, coagulase neg.
staphylococcus Isolate_CFU/ML
43. Number of
antibiotics
Numb_of_AB Number of antibiotics corresponding to
the listed antibiotic in VAE surveillance
protocol
0 – no antibiotic
1- one antibiotic
2 – 2 antibiotics
3- 3 antibiotics
4- 4 antibiotics
5- 5 antibiotics
6- 6 antibiotics
Manually determine for
VAE (IVAC)
44. Ventilator-
associated event
(VAE)
VAE Number of episodes identified as VAE
and not
0- No
1- Yes
45. Ventilator-
associated event
(VAE)
VAE_categories Number of identified as VAE (VAC,
IVAC, PoVAP - meeting the VAE
algorithm
0=No
1. VAC
2. IVAC
3. PoVAP
Manually calculated/
determined using new
CDC VAE surveillance
tool CDC VAE calculator
version 4.0 then used to
confirmation.
46. ETT type Cuffed or uncuffed Whether the ETT is cuffed or not 0- No
1- Yes
47. Overall ETT cuffed
or not
Overall_ETTcuffed_or
_not
Overall Yes (1) is determined if the “yes”
found in the majority of MV days of every
episodes of MV or if the “yes” are equal
with “no” in MV days – counted as Yes
(1)
0- No
1- Yes
48. Reintubation
episodes
Reintubationepisodes The presence of any reintubation was
taken place or not.
0- No
1- Yes
Risk factor_ raw data
226
49. Overall
Reintubation
episodes_for risk
factor analysis
Overall_Reintubatione
pisodes
If any event of reintubation took place in
each days of every episodes (1) consider
as yes and (0) is no/nil.
0- No
1- Yes
Risk factor – will use to
further analysis
50. Routes of intubation Routes_Intubation Routes of endotracheal intubation whether
nasal or oral or tracheosotomy
1- Nasal
2- Oral
3- Tracheostomy
Risk factor_ raw data
51. Overall routes of
intubation
Overall_Routes_Intuba
tion
Determined as with Nasal tube if the
patient has the longest (MV days) on
nasal tube=coded as 1
Determined as with Oral tube if patient
has the longest (MV days) on Oral
tube=coded as 2
Determined as with tracheostomy tube if
the patient has the longest (MV days) on
tracheostomy tube after being intubated
with nasal or oral tube =coded as 3
If the same total numbers of MV days –
consider the last changed/ the newest type
of tube to be overall
1- Nasal
2- Oral
3- Tracheostomy
Risk factor – will use to
further analysis
52. Highest sedation
scores
Highestsedationscore -3 – unresponsive/paralysed
-2 – response to noxious stimuli
-1 – response to gentle touch/voice
0 – awake & able to calm
+1 – restless & difficult to calm
+2- agitated
-3
-2
-1
0
+1
+2
Risk factor_ raw data
53. Overall Highest
sedation score
OverallHighestsedation
score
The score is counted as overall if the score
had the majority numbers of MV days in
episode/s of MV. If the score found in
equal number of MV days; the score on
the last day will be taken as overall
Scores Risk factor – will use to
further analysis
54. Sedation category Sedation categories Score above were recode into nominal (3
categories)
Recoded into 3 categories
1- Deep Sedation
(Score of -2 to -3)
227
2 – Light Sedation
(Score of -1 to +1)
3- Agitated
(Score of +2)
55. Paralytic agent Paralyticagent The administered of any paralytic drugs
for each MV days
yes - if the infusions were given> 4 hours
of infusion or > 4 times bolus dose
no – other than the criteria above
0- No
1- Yes
Risk factor_ raw data
56. Overall Paralytic
agent
Overall_Paralyticagent Overall Yes (1) is determined if the “yes”
found in the majority of MV days of every
episodes of MV or if the “yes” are equal
with “no” in MV days – counted as Yes
(1)
0- No
1- Yes
Risk factor – will use to
further analysis
57. Gastrointestinal
prophylaxis
GIprophylaxis The administered of any GI prophylaxis
drugs for each MV days
0- No
1- Yes
Risk factor_ raw data
58. Overall GI
prophylaxis
OverallGiprophylaxis Overall Yes (1) is determined if the “yes”
found in the majority of MV days of every
episodes of MV or if the “yes” are equal
with “no” in MV days – counted as Yes
(1)
0- No
1- Yes
Risk factor – will use to
further analysis
59. Steroid presence Steroid The presence of steroid for each MV days 0- No
1- Yes
Risk factor_ raw data
60. Overall steroid
presence
Overall_steriod The presence of any steroid even if only
one day of MV- yes
0- No
1- Yes
61. Blood transfusion Blood transfusion The presence of blood transfusion for
each MV days
0- No
1- Yes
62. Overall blood
transfusion
Overall_BT The presence of any steroid even if only
one day of MV- yes
0- No
1- Yes
63. Nasogastric
(NG)tube presence
Nasogastrictubepresen
ce
The presence of NG tube for each MV
days
0- No
1- Yes
Risk factor_ raw data
64. Overall Nasogastric
tube presence
OverallNasogastrictube
presence
Overall Yes (1) is determined if the “yes”
found in the majority of MV days of every
0- No
1- Yes
Risk factor – will use to
further analysis
228
episodes of MV or if the “yes” are equal
with “no” in MV days – counted as Yes
(1)
65. X-Rays with disease X-Ray with disease Yes - if radiological report (in each MV
days) distinguished the findings as: 2 or
more serial X-Rays with 1 of the
following:
new or progressive and persistent
infiltrate
consolidation
cavitation
pneumatoceles, in < 1 y.o
No – if none of radiological report (in
each MV days) found as the list of X-
Rays findings and no X-Ray done on the
particular days
NA- if patient without disease
0- No
1- Yes
3- NA
Requirement for CDC
PNEU/VAP surveillance
tool
66. X-Rays without
disease
X-Ray without disease Yes - if radiological report (in each MV
days) distinguished the findings as: 1 or
more serial X-Rays with 1 of the
following
new or progressive and persistent
infiltrate
consolidation
cavitation
pneumatoceles, in < 1 y.o
No – if none of radiological report (in
each MV days) found as the list of X-
Rays findings and no X-Ray done on the
particular days
NA- if patient with disease
0- No
1- Yes
3- NA
Requirement for CDC
PNEU/VAP surveillance
tool
67. Worsening gas
exchange
[(Increased oxygen
SpO2 < 94% Yes- if any of SpO2 readings <94% in a
day- for every MV days
No
0- No
1- Yes
2- No value
Requirement for CDC
PNEU/VAP surveillance
tool especially for infant
229
requirement, or
increased
ventilation
demand)] (i)FiO2
(ii) PEEP or iii)
Spo2 <94%
<1y.o apart from
worsening oxygenation
SpO2 <94%
Also, for children >1y.o
68. Temperature
instability
Temp<36oC_or>38oC Yes- if any of temperatures was 36oC_ or
>38oC in a day – for every MV days.
No
0- No
1- Yes
2- No value
Requirement for both CDC
PNEU/VAP surveillance
tool and
VAE (IVAC- 2nd tier) [new
CDC VAE surveillance
tool]
69. Tachypnea
RR_aged_based Defined as [(> 75 breath/min –premature
infants born at <37 wks & until 40th wks;
>60bpm=<2 months old; >50bpm= 2-12
months old; >30bpm=children >1 yr.
old)].
Yes – if any of the respiration rates in a
day of each MV days met the tool based
on age classifications as above.
No
0- No
1- Yes
2- No value
Requirement for CDC
PNEU/VAP surveillance
tool
70. Bradycardia or
tachycardia
HR<100or>170_infant
<1yr
Defined as HR <100 or >170 beats /min
Yes- if any of HR reading was <100 or
>170 beats/min in a day – for every MV
days.
No
NA (for infant >1yr. old)
0- No
1- Yes
2- No value
3- NA for patients >1 y.o.
Requirement for CDC
PNEU/VAP surveillance
tool
Applicable for infant <1y.o
only
71. Changes in
character of the
sputum
changesputumcharac Defined as change in sputum character [(i)
colour; (ii) consistency; (iii) quantity
Yes
No
Considering this character first:
Sputum colour: yellow, creamy, green,
brown
Sputum consistency: thick
0- No
1- Yes
Requirement for CDC
PNEU/VAP surveillance
tool
230
Sputum quantity: large, copious, moderate
72. Leukopenia or
Leukocytosis
WBCC (≤4K or ≥15K) Yes- if any of WBCC results was ≤4K or
≥15K in a day – for every MV days.
No
0- No
1- Yes
2- No value/missing/no sample
sent
Requirement for both CDC
PNEU/VAP surveillance
tool and
VAE (IVAC- 2nd-Tier)
[new CDC VAE
surveillance tool]
After gathered all data &
the VAC confirmed;
second screening done for
IVAC determination as
≥12K is required
73. Microbiological
consideration of
respiratory
secretions
Gram_staning Recorded for
ETA
BAL
Pleural fluid
Lung tissue biopsy
Gram+ve, Gram-ve, no growth, no sample
sent :(cocci, bacilli or other) (washing
BAL or BAL) (Pleural fluid RT/LT)
(Lung tissue biopsy)
Requirement for both CDC
PNEU/VAP surveillance
tool
and VAE (possible VAP
(Tier 3) [new CDC VAE
surveillance tool]
74. Corresponding
values to
quantitative
threshold values for
cultured specimens
used in diagnosis of
pneumonia
Colony_forming_unit Documented as
Epi – scant, 1+ -4+
Leu- scant, 1+ - 4+
Sq. epi
Erythrocyte
WBC’s ___x10^6/L; RBC’s___x10^6/L;
Polymorphs___%
Contains clots, blood stained
Requirement for both CDC
PNEU/VAP surveillance
tool
and VAE (possible VAP
(Tier 3) [new CDC VAE
surveillance tool]
75. Name of the
organism with
quantitative
threshold values for
cultured specimens
Organisms isolated No organism seen, normal respiratory
flora, organism name (scant, 1+ -4+), no
fungi isolate, coagulase neg.
staphylococcus Isolate_____CFU/ML
Requirement for both CDC
PNEU/VAP surveillance
tool
and VAE (possible VAP
(Tier 3) [new CDC VAE
surveillance tool]
231
used in diagnosis of
pneumonia
76. Ventilator-
associated
pneumonia (VAP)
VAP Number of identified as VAP case-
meeting the CDC tool from each
experimental unit (from 120 episodes of
MV)
0- No
1- Yes
Manually determined
based on CDC PNEU/VAP
surveillance tool
77. Hand hygiene
practise
HH % of compliance by month (June – Dec
2017)
%
78. Frequency of oral
hygiene practises
MouthCareF Is referred to the frequency of mouth care
was performed for every MV days
Frequency of MC performance
(count)
Raw data
79. Overall OH
Frequency
OverallOHFrequency Average of the frequency to MV days of
every episodes of MV
1 average value recorded
represent each episode of MV
Further analysis required to
determine the compliance
rate based PICU Oral
Hygiene protocol (<6
months without teeth and >
6 months with teeth)
80. Adhered to 12 -
hourly OH
assessment
Adhered to 12- hourly
OH assessment
Adhered to 12 -hourly oh assessment in
every MV days
0- No
1- Yes
81. Overall adhered to
12- hourly OH
assessment
Overall_adhered to 12-
hourly OH assessment
Counted as the majority of yes/no in MV
days – If equal number of Yes/ No found
in one episode of MV – considered as Yes
0- No
1- Yes
82. Adhered to
appropriate OH
Age appropriate OH Adhered to age appropriate OH material <
than six months and > than 6 months as
per PICU oral hygiene protocol for every
MV days
0- No
1- Yes
83. Overall adhered to
appropriate OH
Overall adhered to
appropriate OH
Counted as the majority of yes/no in MV
days – If equal number of Yes/ No found
in one episode of MV – considered as Yes
0- No
1- Yes
84. Frequency of ET
suctioning practises
SuctioningF Is referred to the frequency of ET suction
was performed for every MV days
Frequency of ET performance
(count)
Raw data
85. Overall suctioning
frequency
Overall_suctioning_fre
quency
Average of the frequency to MV days of
every episodes of MV
1 average value recorded
represent each episode of MV
Further analysis required to
determine the compliance
rate based on PICU
232
minimum standard of
suctioning frequency (4
times/ 24 hours)
86. Frequency of cuff
pressure checks
practises
CuffpressurechecksF Is referred to the frequency of cuff
pressure check was performed for every
MV days
Frequency of Cuff Pressure
performance (count)
0 is recorded as uncuffed or
ordered as deflated
Raw data
87. Overall cuff
pressure frequency
Overallcuffpressure_fr
equency
Average of the frequency to MV days of
every episodes of MV
1 average value recorded
represent each episode of MV
Further analysis required to
determine the compliance
rate based on PICU
standard (at least 12 hourly
or twice daily)
88. Adhered to 12
hourly cuff pressure
checks
Adhered to 12 hourly
cuff pressure checks
Adhered to 12 hourly cuff pressure checks
in every MV days
0- No
1- Yes
3 – NA (no cuff or ordered as
deflated)
89. Overall adhered to
12 hourly cuff
pressure checks
Overall_adhered to 12
hourly cuff pressure
checks
Counted as the majority of yes/no in MV
days – If equal number of Yes/ No found
in one episode of MV – considered as Yes
0- No
1- Yes
3 – NA (no cuff or ordered as
deflated)
90. Maintain to cuff
pressure limits
Maintain to cuff
pressure limits
Cuff pressure is maintained min 10 -
15/20cm H20 max
0- No
1- Yes
3- NA (uncuffed/ deflated)
91. Overall Maintain
cuff pressure limits
Overall_Maintain cuff
pressure limits
Counted as the majority of yes/no in MV
days – If equal number of Yes/ No /NA
found in one episode of MV – considered
as Yes
0- No
1- Yes
3 – NA (no cuff or ordered as
deflated)
92. Frequency of head
of bed elevation
practises
HeadofelevationF Is referred to the frequency of head of bed
elevation care was performed in a day for
every MV days
Frequency of HOB performance
(count)
Raw data
93. Overall HOB
frequency
Overall_HOB_frequen
cy
Average of the frequency to MV days of
every episodes of MV
1 average value recorded
represent each episode of MV
Further analysis required to
determine the compliance
rate based on PICU
standard
233
94. Adhered to HOB
degree
Adhered to HOB
degree
Is keeping HOB 15-30 degree every MV
days
0- No
1- Yes
Raw data and later will
compared with unit
standard (24 times/ 24
hours)
95. Overall adhered to
HOB degree
Overall_Adhered to
HOB degree
Counted as the majority of yes/no in MV
days – If equal number of Yes/ No found
in one episode of MV – considered as
Yes. (Is keep HOB 15-30 degree)
0- No
1- Yes
96. Frequency of
ventilator circuits
practises
VentilatorcircuitsF Is referred to the frequency of ventilator
circuits was performed for every MV days
(Unit standard: 24 times/ 24 hours)
Frequency of Ventilator Circuit
performance (count)
Raw data
97. Overall Vent circuit
check_ frequency
Overall_Vent_cir_chec
k_frequency
Average of the frequency to MV days of
every episodes of MV
1 average value recorded
represent each episode= of MV
Further analysis required to
determine the compliance
rate based PICU standard
98. Enteral feeding
started within 24
hours of admission
Enteral feeding started
within 24 hours of
admission
The enteral feeding is started within 24
hours of admission
0- No
1- Yes
Descriptive analysis
234
Appendix L Parents information sheet and consent form for prospective study
235
236
237
Appendix M Approval of waiver of consent for prospective study