microbiology and immunology of otitis media · otitis media chinh chung ngo master of science...

Microbiology and Immunology of

Otitis Media

Chinh Chung Ngo

Master of Science (Genetics)

Submitted in fulfilment of the requirements of the degree of

Doctor of Philosophy

School of Medical Science

Griffith University

April 2016

iii

Statement of Originality

This work has not previously been submitted for a degree or diploma in any

university. To the best of my knowledge and belief, the thesis contains no

material previously published or written by another person except where due

reference is made in the thesis itself.

_____________________________

Chinh Chung Ngo

iv

Abstract

Background

Over 80% of children experience acute otitis media (AOM) at least once during

childhood, with 10-30% of children experiencing recurrent episodes of AOM

(RAOM) or persistence of fluid in the middle ear (ME) lasting greater than 3

months which is defined as chronic otitis media with effusion (COME). Microbial

infection remains the main cause of AOM, however the causes of RAOM and

COME are not fully understood. Heavy loads of microbial colonisation and

bacterial/viral interaction in the upper respiratory tract (URT) contribute to OM

pathogenesis. It is currently unclear whether compromised host immune system

responses have a significant role in RAOM/COME. Bacterial biofilms within the

ME may contribute to otopathogen persistence and recurrent infection and/or

inflammation in these children.

Aims

This study identified the predominant bacterial and viral otopathogens within the

URT and ME, of children undergoing ventilation tube insertion (VTI) for

RAOM/COME. Bacterial persistence within the ME, in the form of biofilms was

also examined.

Bacteria and viruses which were identified within the URT of children with

RAOM/COME were compared to a control group of children who were

undergoing adenoidectomy for treatment of adenoidal hypertrophy (AH) and/or

obstructive sleep apnoea (OSA).

Local (saliva, middle ear effusion (MEE)) and systemic (serum) innate and

adaptive immune responses were examined in children with and without

RAOM/COME. Specific antibody levels to bacterial proteins as well as cytokine

levels were determined. Systemic immuno-gene expression was compared

between children with RAOM/COME and AH/OSA.

v

Methods

URT samples, including adenoids, nasopharyngeal swabs (NPS), MEE and ME

biopsies were collected from 23 urban children with RAOM/COME between

2008 and 2011 (OM1 cohort). Additional samples; nasal swabs, bloods and

saliva were collected from 20 urban children with RAOM/COME (OM2 cohort)

and 17 urban children with AH/OSA (the control cohort) in 2015.

The presence of S. pneumoniae (Spn), H. influenzae (Hi) and M. catarrhalis

(Mcat) within the URT were identified using bacterial culture and PCR methods.

Bacterial biofilm within the middle ear was identified by LIVE/DEAD viability

staining and fluorescence in situ hybridisation (FISH).

Viruses were detected within the URT, using PCR for common respiratory

viruses, including human rhinovirus (HRV), respiratory syncytial virus (RSV) and

adenovirus (ADV) and up to 14 other viruses.

Antibodies (IgA and IgG) against protein vaccine candidates of Spn and Hi were

analysed in serum, saliva and MEE in the OM2 cohort, and serum and saliva in

the control cohort. Ten cytokines were examined in the serum and MEE from

the OM2 cohort and serum from the control cohort. Regulatory T (Treg) cells

were examined from peripheral blood mononuclear cells (PBMC) in children

from the OM2 and control cohorts.

Nanostring technology and Ingenuity Pathway Analysis (IPA) software

facilitated examination of 579 immuno-genes in PBMC of children with/without

RAOM/COME.

Results

Bacterial otopathogens were most frequently detected in adenoids, then NPS,

nasal and MEE samples from children with RAOM/COME. H. influenzae was

the predominant bacteria within the URT of children with/without RAOM/COME.

Mcat prevalence increased whilst Spn prevalence decreased, pre and post

PCV13 introduction (Chapter 3).

vi

Viruses were most frequently detected within adenoid samples for both

RAOM/COME and control children. HRV was the predominant virus detected in

all URT locations (Chapter 4).

Local and systemic IgA and IgG production against Spn and Hi protein antigens

did not differ between children with/without RAOM/COME. Children with

RAOM/COME had measurable cytokines in their MEE and this was higher in

the presence of pathogens. Comparable frequencies of Treg cells were

observed between children with and without RAOM/COME (Chapter 5).

Normal systemic immuno-gene expression, responsible for immune response

regulation against bacterial infection was not significantly different in children

with/without RAOM/COME. Two unique genes HLA-DQA1 and HLA-DQB1

genes were differentially expressed in children with/without RAOM/COME

(Chapter 6).

Conclusions

Hi and HRV remain the predominant otopathogens identified in the MEE of

children with RAOM/COME. PCV13 vaccination may have altered the relative

prevalence of Mcat and Spn and facilitated serotype replacement in the URT.

Ongoing surveillance of these microbes is required to guide vaccine

development.

Immune responses to microbial infections are not compromised in children with

RAOM/COME although 2 unique genes HLA-DQA1 and HLA-DQB1 require

further investigation for their role in OM pathogenesis. Overall, the use of multi-

pathogen vaccines to reduce or prevent OM remains feasible.

vii

Acknowledgements

First and foremost, I would like to thank my supervisors Prof. Allan W. Cripps,

A/Prof. Helen M. Massa and Dr. Ruth B. Thornton, for their continuous support

during my PhD study. They inspired me greatly to work on this project. I would

not have progressed nearly as far as I did, without their support, help and

guidance.

I would like to acknowledge all children and their families for unconditionally

participating in this project.

I would also like to acknowledge the surgeons and nurses, including A/Prof.

Christopher Perry, Dr. Michael Nissen, Dr. Andrew Chang, Dr. Richard Barr,

Nicholas O’Neill, Elisabeth Hodge and Stephen Lambert at the Royal Children’s

Hospital in Brisbane and Dr. Brent McMonagle, Dr. David McGuire, Jo Tier,

Jenny, Melanie and Jillian at Pindara Hospital in the Gold Coast for participating

in these studies. The surgeons and theatre teams were vital to the performance

of the study and their ongoing support and interest in this research was very

much welcomed.

I would like to say thank you to Prof. Paul Rigby and Alysia Buckley at CMCA,

University of Western Australia, for the training I received and for making me

feel comfortable with microscopy. To Prof. Peter Richmond and the staff of

Vaccine Trials Group, A Telethon Kids Institute, for making me feel welcome

during my time in Western Australia.

I would like to make a special thank you to Prof. Theo Sloots, Dr. Jane Gaydon,

Rebecca Holding, Rebecca Rockett and A/Prof. Seweryn Bialasiewicz at the

AID research centre, for their collaboration with my viral studies. To Dr. Dale

Thorley and Dr. Petra Derrington at Gold Coast University Hospital, Shannen

Fentiman and Mario Mitov at Griffith University for their collaboration with the

bacterial part of my research.

A huge thank you to Dr. Nicholas West and Dr. Amanda Cox for their support

and assistance during the project.

viii

A big thank you to Dr. Benjamin Duell, Dr. Malgorzata (Gosia) Skalska, Dr.

Helen Irving-Rodgers, Dr. Ping Zhang, Dr. Alison Carey, Dr. Ruth Freeman, Dr.

Cameron Flegg, Dr. Vinod Gopalan, Dr. Pauline Young, A/Prof. Jing Sun, Dr.

Ekua Brenu and Teilah Huth at Griffith University for their assistance with

specific techniques that I have used for the project.

To HDR students and post-docs from my room at G40, especially Moshen

Sohrabi, thank you for your friendship, interesting talks and fun times.

I would like to acknowledge the Vietnam International Education Development-

Griffith University co-funded PhD scholarship, Griffith University conference

travel grant scholarship, Griffith University Heart Foundation scholarship, and

CAPRS scholarship for funding my studies and travels during my studies. I

would also like to acknowledge the Royal Children's Hospital Foundation, Gold

Coast Hospital Foundation and Financial Markets for Children Foundation for

funding this project.

Last but not least, thanks to my family, mom, dad (in heaven), sisters and

brothers, whose support has been an important and comforting part of my life.

ix

Publications and Presentations

Publications

Published publications

Ngo CC, Massa HM, Thornton RB, Cripps AW. 2016. Predominant bacteria

detected from the middle ear fluid of children experiencing otitis media: a

systematic review. PLoS ONE 11(3): e0150949.

Publications in preparation

Ngo CC, Thornton RB, Rockett R, Gaydon J, Holding R, Bialasiewicz S, Sloots

T, Cripps AW, Massa HM. Predominant bacteria and viruses in urban children

with RAOM/COME in Australia.

Ngo CC, Thornton RB, Massa HM, Cripps AW. Systemic and local IgG and IgA

antibodies to pneumococcal and Haemophilus influenzae protein antigens in

children with RAOM/COME.

Ngo CC, Thornton RB, West N, Cox A, Massa HM, Cripps AW. Systemic

immune-gene expression in children with RAOM/COME.

Conference presentations

Ngo C, Rockett R, Sloots T, Thornton R, Massa H, Cripps A. 2015.

Predominant bacteria and viruses in urban children undergoing ventilation tube

insertion due to otitis media in Australia. 18th International Symposium on

Recent Advances in Otitis Media. National Harbor, Maryland, USA. (Oral

presentation).

Ngo CC, Rockett RJ, Sloots TP, Thornton RB, Massa HM, Cripps AW. 2015.

Predominant bacteria in the upper respiratory tract of urban children with otitis

media in South East Queensland. 2015 Gold Coast Health and Medical

Research Conference. Gold Coast, Queensland, Australia. (Poster

presentation).

Ngo C, Thornton RB, Cripps AW, Massa HM. 2013. Otopathogenic biofilms in

urban children undergoing ventilation tube insertion. 2013 Gold Coast Health

x

and Medical Research Conference. Gold Coast, Queensland, Australia. (Poster

presentation).

x

Table of Contents

Statement of Originality ................................................................................... iii

Abstract ............................................................................................................. iv

Acknowledgements ......................................................................................... vii

Publications and Presentations ...................................................................... ix

Table of Contents .............................................................................................. x

List of Figures ............................................................................................... xviii

List of Tables ................................................................................................. xxii

Abbreviations ............................................................................................... xxvi

Chapter 1: Literature review ................................................................... 1

1.1 Systematic review ........................................................................... 2

1.2 Otitis media ...................................................................................... 4

1.2.1 Acute otitis media ...................................................................... 4

1.2.2 Otitis media with effusion .......................................................... 4

1.3 Pathogenesis of OM ........................................................................ 5

1.4 Microbiology of OM ......................................................................... 8

1.4.1 Role of bacteria in OM .............................................................. 8

1.4.1.1 Three main otopathogens ................................................. 8

1.4.1.2 The origin of bacterial otopathogens .............................. 10

1.4.2 Role of viruses in OM .............................................................. 11

1.4.3 Biofilms in OM ......................................................................... 12

1.5 Immunology of OM ........................................................................ 14

1.5.1 Cytokines in OM ...................................................................... 14

1.5.2 Serum antibody responses in OM ........................................... 15

1.5.3 Mucosal antibody responses in OM ........................................ 17

1.5.4 Regulatory T cells in OM ......................................................... 18

1.6 Systemic gene expression in OM ................................................. 18

1.7 OM treatment ................................................................................. 20

xi

1.7.1 Antibiotics ................................................................................ 20

1.7.2 Surgery ................................................................................... 22

1.8 Vaccines and OM ........................................................................... 23

1.8.1 Bacterial vaccines ................................................................... 24

1.8.2 Viral vaccines .......................................................................... 25

1.9 Significance and aims of the project ........................................... 27

Chapter 2: General methods ................................................................ 30

2.1 Study populations and recruitment ............................................. 31

2.1.1 Pilot project: OM1 cohort ......................................................... 31

2.1.2 Primary study: OM2 cohort and control cohort ........................ 32

2.2 Sample collection .......................................................................... 35

2.2.1 Saliva ...................................................................................... 35

2.2.2 Samples collected under anaesthesia..................................... 35

2.2.2.1 Blood .............................................................................. 35

2.2.2.2 Nasopharyngeal swab .................................................... 35

2.2.2.3 Nasal swab ..................................................................... 36

2.2.2.4 Middle ear effusion ......................................................... 36

2.2.2.5 Middle ear mucosa ......................................................... 36

2.2.2.6 Adenoid tissue ................................................................ 37

2.3 Bacterial culture ............................................................................ 37

2.3.1 Clinical samples ...................................................................... 37

2.3.2 Antibiotic susceptibility assay .................................................. 38

2.3.3 Standard bacterial control samples ......................................... 39

2.4 Viral PCR ........................................................................................ 40

2.4.1 Nucleic acid extraction for viral detection ................................ 40

2.4.2 Viral detection ......................................................................... 41

2.5 Bacterial PCR ................................................................................. 41

2.5.1 Optimisation of DNA extraction from gram-positive bacteria ... 41

2.5.2 DNA extraction for bacterial detection ..................................... 44

2.5.3 Bacterial detection .................................................................. 44

2.5.4 Bacterial load using quantitative PCR ..................................... 45

xii

2.6 S. pneumoniae serotyping ............................................................ 46

2.7 H. influenzae typing ....................................................................... 48

2.8 LIVE/DEAD staining of middle ear effusion................................. 50

2.8.1 Testing reagent with standard bacteria ................................... 50

2.8.2 Sample staining ...................................................................... 53

2.9 Fluorescent in situ Hybridisation ................................................. 53

2.9.1 Specificity of probes ................................................................ 53

2.9.2 FISH of middle ear biopsy ....................................................... 60

2.10 Serum collection ............................................................................ 61

2.11 IgA and IgG assays ....................................................................... 61

2.12 Cytokine assays ............................................................................ 62

2.13 Peripheral blood mononuclear cell isolation .............................. 62

2.14 Regulatory T cell assay ................................................................. 63

2.14.1 Thawing cryopreserved PBMC ............................................... 63

2.14.2 Cell staining for flow cytometry ............................................... 66

2.14.3 Flow cytometry analysis .......................................................... 68

2.14.3.1 Setting up and controls ................................................... 68

2.14.3.2 Gating strategy ............................................................... 70

2.15 Gene expression assay ................................................................. 72

2.15.1 Cell lysate preparation ............................................................ 72

2.15.2 RNA hybridisation ................................................................... 72

2.15.3 Reaction analysis .................................................................... 72

2.15.4 Data analysis .......................................................................... 73

2.16 Statistical analysis ........................................................................ 74

Chapter 3: Bacterial otopathogens in the upper respiratory tract

and middle ear of children with and without RAOM/COME in

South East Queensland .......................................................................... 76

3.1 Introduction.................................................................................... 77

3.2 Methods .......................................................................................... 79

3.3 Results ........................................................................................... 81

xiii

3.3.1 Study population ..................................................................... 81

3.3.2 Bacterial carriage .................................................................... 83

3.3.2.1 Bacteria identified within the adenoids of children with

RAOM/COME compared to control ................................ 83

3.3.2.2 Bacterial detection in the nasopharynx of children with

RAOM/COME compared to controls ............................... 87

3.3.2.3 Bacterial detection from the nasal vestibule of children

with RAOM/COME compared to controls ....................... 89

3.3.2.4 Bacterial detection from the MEE of children with

RAOM/COME compared to controls ............................... 90

3.3.3 Relationship of bacterial detections in sampling sites of the

upper respiratory tract ............................................................. 93

3.3.4 Co-occurrence of bacterial otopathogens in the upper

respiratory tract of children with and without RAOM/COME .... 97

3.3.5 Bacterial load in the nasal, nasopharynx and adenoids of

children with and without RAOM/COME ................................. 99

3.3.6 Pneumococcal serotypes in the upper respiratory tract of

children with and without RAOM/COME ............................... 101

3.3.7 Hi typing for samples from the upper respiratory tract of children

with and without RAOM/COME ............................................. 102

3.3.8 Antibiotic susceptibility of bacterial isolates from the upper

respiratory tract of children with and without RAOM/COME .. 103

3.3.8.1 S. pneumoniae ............................................................. 103

3.3.8.2 H. influenzae ................................................................. 103

3.3.8.3 M. catarrhalis ................................................................ 104

3.3.9 Bacterial biofilms and intracellular bacteria in MEE of children

with RAOM/COME ................................................................ 105

3.3.10 Bacterial biofilms on middle ear biopsies of children with

RAOM/COME ....................................................................... 109

3.4 Discussion ................................................................................... 111

xiv

Chapter 4: Viruses identified within the upper respiratory tract

of urban children with and without RAOM/COME in South East

Queensland ............................................................................................... 119

4.1 Introduction.................................................................................. 120

4.2 Methods ........................................................................................ 121

4.3 Results ......................................................................................... 123

4.3.1 Study population ................................................................... 123

4.3.2 Viral carriage ......................................................................... 125

4.3.2.1 Viral identification from the adenoids of children with and

without RAOM/COME ................................................... 125

4.3.2.2 Viral identification from the nasopharynx of children with

and without RAOM/COME ............................................ 128

4.3.2.3 Viral identification from the nasal vestibule of children with

and without RAOM/COME ............................................ 129

4.3.2.4 Viral detection in the middle ear of children with

RAOM/COME ............................................................... 130

4.3.3 Co-detection of specific viruses in different locations of the

upper respiratory tract of the same child from the OM and

control cohorts ...................................................................... 133

4.3.4 Co-detection of common viruses in the nasopharynx and

adenoids of children with and without RAOM/COME ............ 136

4.3.5 Co-detection of common viruses and bacterial otopathogens in

the upper respiratory tract of children with and without

RAOM/COME ....................................................................... 140

4.4 Discussion ................................................................................... 146

Chapter 5: Immune responses in urban children with and

without RAOM/COME in South East Queensland ....................... 153

5.1 Introduction.................................................................................. 154

5.2 Methods ........................................................................................ 155

5.3 Results ......................................................................................... 157

xv


5.3.2 Serum immune responses in children with and without

RAOM/COME ....................................................................... 159

5.3.2.1 Pneumococcal protein specific IgG and IgA antibody

levels in serum of children with and without

RAOM/COME ............................................................... 159

5.3.2.2 Hi protein specific IgG and IgA antibody levels in serum of

children with and without RAOM/COME ....................... 161

5.3.3 Serum antibody levels according to nasopharyngeal

colonisation in children with and without RAOM/COME ........ 163

5.3.3.1 Serum antibody titres to pneumococcal proteins in children

with and without Spn colonisation ................................. 163

5.3.3.2 Serum antibody titres to Hi proteins in children with and

without Hi colonisation .................................................. 165

5.3.4 Serum antibody levels based on Spn and Hi in the middle ear of

children with RAOM/COME ................................................... 167

5.3.5 Mucosal antibodies against pneumococcal and Hi proteins in


5.3.5.1 Pneumococcal specific IgA and IgG in the MEE of children

with RAOM/COME ........................................................ 169

5.3.5.2 IgG and IgA to Hi protein antigens in the MEE of children

with RAOM/COME ........................................................ 171

5.3.5.3 Antibody titres in saliva of children with and without

RAOM/COME ............................................................... 173

5.3.5.4 Correlation in levels of systemic (serum) and local (MEE

and saliva) antibodies in children with and without

RAOM/COME ............................................................... 175

5.3.6 Serum and MEE cytokine titres in children with and without

RAOM/COME ....................................................................... 177

5.3.6.1 Correlation of cytokine levels in MEE samples ............. 178

5.3.6.2 Cytokine titres in relation to bacterial and viral pathogen

presence in the middle ear ........................................... 179

xvi

5.3.7 Regulatory T cells in peripheral blood mononuclear cells in


5.4 Discussion ................................................................................... 183

Chapter 6: Systemic immune gene expression in urban children

with and without RAOM/COME in South East Queensland ..... 187

6.1 Introduction.................................................................................. 188

6.2 Methods ........................................................................................ 189

6.3 Results ......................................................................................... 192


6.3.2 Gene expression between OP and NOP children ................. 193

6.3.2.1 Canonical pathways of gene expression in OP children 195

6.3.2.2 Biological functions and networks of gene expression in

OP children ................................................................... 197

6.3.3 Differences in gene expression amongst OP children with MEE

positive and negative for bacterial otopathogens .................. 201

6.3.3.1 Canonical pathways of gene expression in OP children

with MEE positive for bacterial otopathogens ............... 205


OP children with MEE positive for bacterial

otopathogens ................................................................ 207

6.3.4 Differences in gene expression amongst OP children with MEE

positive and negative for any tested virus ............................. 213


with MEE positive for any tested viruses ...................... 216


OP children with MEE positive for any tested virus ...... 219

6.4 Discussion ................................................................................... 227

Chapter 7: General discussion .......................................................... 232

Bibliography ............................................................................................... 242

xvii

Appendix A: Bacterial otopathogens: systemic review ................................... 266

Appendix B: Media and solutions ................................................................... 293

Appendix C: List of genes in Nanostring analysis ........................................... 298

xviii

List of Figures

Chapter 1

Figure 1-1: Normal middle ear, acute otitis media and otitis media with

effusion ............................................................................................................... 5

Figure 1-2: Key factors of pathogenesis of otitis media ..................................... 6

Figure 1-3: The general process of bacterial biofilm formation ........................ 14

Chapter 2

Figure 2-1: Samples collected from the OM1 cohort for microbiological

analyses ........................................................................................................... 33

Figure 2-2: Samples collected from the OM2 and control cohort patients for

microbiological and immunological analyses .................................................... 34

Figure 2-3: Plating sample for semi-quantitative of bacteria detection ............ 38

Figure 2-4: The impact of pre-treatment of bacteria with different enzymes or

TE on the concentration of extracted DNA ....................................................... 43

Figure 2-5: Representative standard curve of quantitative PCR for bacterial

load analysis. .................................................................................................... 46

Figure 2-6: Representative images of positive and negative pneumococcal

antigen and antibody reactions of Quellung test............................................... 48

Figure 2-7: Scheme for typing H. influenzae strains by PCR ........................... 50

Figure 2-8: Pure culture of ATCC H. influenzae was stained with LIVE/DEAD

BacLight reagents ............................................................................................ 52

Figure 2-9: Mixture of all standard bacteria were stained with Hoechst 33342

for DNA, NONEUB338-AF488 probe (non-bacterial probe) as a negative control

and EUB338-AF546 probe, which is a universal bacterial probe, as a positive

control .............................................................................................................. 56

Figure 2-10: Mixture of standard bacteria without Spn were stained with

Hoechst 33342, Spn-AF488 probe, EUB338-AF546 probe and Hi-AF633 probe

to evaluate the specificity of Spn-AF488 probe ................................................ 57

xix

Figure 2-11: Mixture of standard bacteria without Hi were stained with Hoechst

33342, Spn-AF488 probe, EUB338-AF546 probe and Hi-AF633 probe to

evaluate the specificity of Hi-AF633 probe ....................................................... 58

Figure 2-12: Mixture of all standard bacteria including Spn and Hi were stained

with Hoechst 33342, Spn-AF488 probe, EUB338-AF546 probe and Hi-AF633

probe to assess multiple probes used in one sample ....................................... 59

Figure 2-13: Serum collection from whole blood using centrifugation .............. 61

Figure 2-14: PBMC isolation from whole blood using Ficoll separation ........... 63

Figure 2-15: Comparison cell viability and viable cell recovery between fresh

PBMC, thawed PBMC with and without resting overnight ................................ 65

Figure 2-16: Overview of Treg cell staining procedure adapted from BD

protocol ............................................................................................................ 67

Figure 2-17: Setting controls and gating for each subset of T lymphocyte using

FMO and isotype .............................................................................................. 69

Figure 2-18: Gating strategy for Treg cells from PBMC ................................... 71

Chapter 3

Figure 3-1: Co-detection of bacterial otopathogens in different locations of the

upper respiratory tract from children with and without RAOM/COME ............... 96

Figure 3-2: Bacterial loads were determined in the NS, NPS and adenoidal

samples from children with and without RAOM/COME using semi-quantitative

bacterial culture and qPCR............................................................................. 100

Figure 3-3: Bacterial microcolonies and biofilms identified in MEE samples . 106

Figure 3-4: Representative image of LIVE/DEAD stained MEE from a child with

a history of RAOM/COME (OM2 Child 32 right ear) showing the complex 3-

dimensional (3-D) morphology of the bacterial biofilms and the nature of

bacterial involvement, live or dead, intracellular, and extracellular within the

biofilm ............................................................................................................. 108

Figure 3-5: Bacterial microcolonies identified in the middle ear biopsy from a

child with RAOM/COME ................................................................................. 110

xx

Chapter 4

Figure 4-1: Co-detection of specific viruses in different locations of the upper

respiratory tract from children with and without RAOM/COME ....................... 135

Chapter 5

Figure 5-1: Geometric mean titres of IgG and IgA to Spn proteins in serum of

children with and without RAOM/COME ......................................................... 160

Figure 5-2: Geometric mean titres of IgG and IgA to Hi proteins in serum of

children with and without RAOM/COME ......................................................... 162

Figure 5-3: Geometric mean titres of IgG and IgA to Spn proteins in serum from

children with and without Spn nasopharyngeal colonisation within the OM and

control cohorts. ............................................................................................... 164

Figure 5-4: Geometric mean titres of IgG and IgA to Hi proteins in serum from

children with and without Hi colonisation in nasopharynx within the OM and

control cohorts ................................................................................................ 166

Figure 5-5: Geometric mean titres for serum IgG and IgA to Spn and Hi

proteins between children with MEE positive or negative for bacteria in the OM

cohort ............................................................................................................. 168

Figure 5-6: Geometric mean IgG and IgA titres to PspA1, PspA2, Ply and CbpA

in MEE from children with RAOM/COME ....................................................... 170

Figure 5-7: Geometric mean IgG and IgA titres against P26, P4, P6 and PD in

MEE from children with RAOM/COME ........................................................... 172

Figure 5-8: Salivary IgG and IgA against Spn and Hi proteins in children with

and without RAOM/COME.............................................................................. 174

Figure 5-9: Cytokines in MEE based on microbial detection ......................... 180

Figure 5-10: Levels of cytokines in MEE positive and negative for specific

bacteria and virus ........................................................................................... 181

Chapter 6

Figure 6-1: Canonical pathways predicted in OP children ............................. 196

xxi

Figure 6-2: Shown is the decreased “cell death of immune cells” function

predicted in OP children ................................................................................. 197

Figure 6-3: Network #1 showed connection amongst identified genes. Green

shapes indicate those genes that were expressed at a lower rate in OP children,

compared to NOP children ............................................................................. 200

Figure 6-4: Three biological functions predicted in children with bacterial

otopathogens detected from their MEE .......................................................... 209

Figure 6-5: The TLR signalling pathway (an innate immune response pathway)

was predicted to be activated in OP children who had virus positive MEE ..... 218

Figure 6-6: “Upregulated” or “increased” expression of genes involved in

increase of “development of lymphocytes” and in decrease of “infection of

mammalia” in children with MEE that was positive for any of the viruses

assessed ........................................................................................................ 223

Supplemental Figure 6-1: Geometric mean count of number of gene expression

of cytokines and TLRs in PBMC of OP and NOP children ............................. 230


in PBMC of OP children with MEE positive and negative for bacteria ............ 231

xxii

List of Tables

Chapter 1

Table 1-1: Risk factors for OM ........................................................................... 7

Table 1-2: Bacterial and viral findings of MEE from children with AOM ........... 12

Chapter 2

Table 2-1: Zone diameter breakpoints to determine specific antibiotic

susceptibility for the 3 main bacterial otopathogens ......................................... 39

Table 2-2: Primers and probes for bacterial detection using real-time PCR .... 45

Table 2-3: Chessboard of Pneumotest kit for pneumococcal serotype

identification based on Quellung reaction ......................................................... 47

Table 2-4: Primers used for PCR typing H. influenzae ..................................... 49

Table 2-5: Probes used to identify bacterial species using FISH ..................... 55

Table 2.6: In situ hybridisation buffer. .............................................................. 55

Table 2-7: Washing buffer appropriate to formamide concentration used in

hybridisation buffer ........................................................................................... 55

Table 2-8: Sample size for each group based on results from previous studies

in Western Australia ......................................................................................... 75

Chapter 3

Table 3-1: Characteristic demographic data for children in each recruitment

cohort enrolled in this study .............................................................................. 82

Table 3-2: Bacterial otopathogens detection from the adenoids, nasopharynx

(NPS), nasal (NS) and middle ear (MEE) of children with RAOM/COME (OM1

and OM2) and children without RAOM/COME (control) ................................... 85

Table 3-3: Bacterial otopathogens identified in the upper respiratory tract of

children fully vaccinated with either PCV7 or PCV13 within the OM and control

groups .............................................................................................................. 86

xxiii

Table 3-4: Individual pathogen detection in the middle ears of children with OM

that had at least one middle ear positive for bacterial otopathogens ................ 92

Table 3-5: Co-occurrence of the three main otopathogens in the nasopharynx

and the adenoids of children from the OM and Control cohorts ....................... 98

Table 3-6: Pneumococcal serotypes in children with and without

RAOM/COME ................................................................................................. 102

Table 3-7: Summary culture and PCR results of 11 MEE samples and FISH

results of 11 MEM samples from children with RAOM/COME ........................ 109

Chapter 4


cohort enrolled in this study ............................................................................ 124

Table 4-2: Viral detection from the adenoids, nasopharynx (NPS), nasal (NS)

and middle ear (MEE) of children with RAOM/COME (OM1 and OM2) and

children without RAOM/COME (control) ......................................................... 127

Table 4-3: Viral results in left and right ears of the same child from 24 children

in the OM1 and OM2 cohorts ......................................................................... 132

Table 4-4: Co-detection of common viruses in the nasopharynx of children from

the OM and control cohorts ............................................................................ 137

Table 4-5: Co-detection of common viruses in the adenoids of children from the

OM and control cohorts .................................................................................. 138

Table 4-6: Microbial results in the upper respiratory tract of children from the


Table 4-7: Co-detection of common viruses and bacterial otopathogens in the

middle ear of children with RAOM/COME ...................................................... 142


nasopharynx of children with and without RAOM/COME ............................... 143


adenoids of children with and without RAOM/COME ..................................... 145

Supplement Table 4-1: Comparison of viral detection amongst different

locations of the upper respiratory tract of children within the OM1, OM2 and

control cohorts. ............................................................................................... 152

xxiv

Chapter 5

Table 5-1: Characteristic demographic data for children in each cohort of this

study ............................................................................................................... 158

Table 5-2: Frequencies of MEE samples with anti-pneumococcal protein IgA

and IgG titres above the limit of detection (LOD) from children in the OM2

cohort ............................................................................................................. 169

Table 5-3: The number of MEE samples having anti-Hi protein IgG and IgA

titres higher than limit of detection (LOD) from children of OM2 cohort .......... 171

Table 5-4: The number of saliva samples having anti Spn and Hi protein IgG

and IgA titres higher than limit of detection (LOD) from both OM and control

cohorts ........................................................................................................... 173

Table 5-5: Correlation in levels of specific antibodies in MEE, saliva and serum

of children from the OM and control cohorts .................................................. 176

Table 5-6: MEE samples with cytokine concentrations which were higher than

the limit of detection (LOD) based on bacterial otopathogen presence in children

from the OM2 cohort ...................................................................................... 177

Table 5-7: Correlation in levels of cytokines in MEE samples from children with

RAOM/COME ................................................................................................. 178

Table 5-8: Proportion of T lymphocyte subsets in PBMC of children from the


Chapter 6

Table 6-1: Demographic data for children in each cohort .............................. 192

Table 6-2: Annotated genes which were differentially expressed (fold change

1.5 or ≤ -1.5) in PBMC between OP and NOP children .................................. 194

Table 6-3: Predicted networks relating differentially expressed genes between

OP and NOP children ..................................................................................... 199

Table 6-4: Genes that were differentially expressed (fold change 1.5 or ≤ -1.5)

in PBMC between children whose MEE was positive or negative for any of the

main bacterial otopathogens .......................................................................... 202

Table 6-5: Canonical pathways predicted in MEE-POS-BAC children ........... 206

xxv

Table 6-6: Biological functions of immune responses predicted in children who

had MEE that were positive for bacterial otopathogens ................................. 207

Table 6-7: Predicted networks relating to differentially expressed genes

between OP children who had MEE which were positive or negative for

bacterial otopathogens ................................................................................... 211

Table 6-8: Genes which are differentially expressed (fold change 1.5 or ≤-1.5)

in PBMC between OP children with positive or negative MEE for viruses

tested ............................................................................................................. 214

Table 6-9: Predicted canonical pathways in OP children with MEE positive for

any of the viruses analysed ............................................................................ 217

Table 6-10: Biological functions of immune responses predicted in PBMC from

OP children with MEE that were positive for any of the viruses assessed ..... 220

Table 6-11: Predicted networks relating to differentially expressed genes

between children who had MEE that were positive or negative for any of the

viruses assessed ............................................................................................ 225

xxvi

Abbreviations

229E 229E coronavirus

% Percent

°C Degrees Celsius

μl Microlitre

Abx Antibiotics

AD Adenoids

ADV Adenovirus

AF Alexa Fluor

AH Adenoids hypertrophy

AU Arbitrary unit

AOM Acute otitis media

ATCC American type culture collection

BAL Bronchoalveolar lavage

CbpA Choline binding protein A

CDC Centers for Disease Control and Prevention

CF Cystic fibrosis

cfu Colony forming units

ChoP Phosphorylcholine

CLSM Confocal laser scanning microscopy

CNS Coagulase negative staphylococci

CO2 Carbon dioxide

COME Chronic otitis media with effusion

CSOM Chronic suppurative otitis media

DNA Deoxyribonucleic acid

EDTA Ethylenediaminetetraacetic acid

ET Eustachian tube

EUB Eubacterial

EV Enterovirus

FISH Fluorescent in situ hybridisation

fM Femtomolar

g (weight) Grams

g (speed) Gravitational force

xxvii

gDNA Genomic DNA

h Hour

HboV Human Bocavirus

HCl Hydrochloric acid

Hh Haemophilus haemolyticus

Hi Haemophilus influenzae

Hib Haemophilus influenzae type b

HKU1 HKU1 coronavirus

HRV Human Rhinovirus

hMPV Human Metapneumovirus

Hpd Haemophilus protein D

IAV Influenza A virus

IBV Influenza B virus

IFN Interferon

Ig Immunoglobulin

IgA Immunoglobulin A

IgG Immunoglobulin G

IgM Immunoglobulin M

IPA Ingenuity Pathway Analysis

KI KI polyomavirus

IL Interleukin

IPA Ingenuity Pathway Analysis

L Litre

LPS Lipopolysaccharide

lytA Autolysin A

Mcat Moraxella catarrhalis

MEE Middle ear effusion

MEF middle ear fluid

MEM middle ear mucosa

mg Milligram

min Minute

ml Millilitre

mm Millimetre

mM Millimolar

xxviii

na Not available

NaCl Sodium chloride

NaOH Sodium hydroxide

NETs Neutrophil extracellular traps

NIP National Immunisation Program

NL63 NL63 coronavirus

nm Nanometre

NPS Nasopharyngeal swab

NS Nasal swab

nt Not able to type

NTHi Nontypeable Haemophilus influenzae

OC43 OC43 coronavirus

OM Otitis media

OME Otitis media with effusion

OSA Obstructive sleep apnea

P26 Protein 26 of H. influenzae



PA Pseudomonas aeruginosa

PAFR Platelet activating factor receptor

PBMC Peripheral blood mononuclear cell

PBS Phosphate buffered saline

PCR Polymerase chain reaction

PCV Pneumococcal conjugate vaccine

PD Protein D of H. influenzae

PFA Paraformaldehyde

pg Picograms

PIV1 Parainfluenza virus 1



Ply Pneumolysin

PspA1 Pneumococcal surface protein A family 1

PspA2 Pneumococcal surface protein A family 2

PVA Poly vinyl alcohol

xxix

qPCR Quantitative polymerase chain reaction

QS Quorum sensing

RAOM Recurrent acute otitis media

RBC Red blood cells

RNA Ribonucleic acid

rpm Revolutions per minute

rRNA Ribosomal RNA

RSV Respiratory Syncytial virus

rxn Reactions

Sau Staphylococcus aureus

SDS Sodium dodecyl sulphate

sIgA Secretory immunoglobulin A

Spn Streptococcus pneumoniae

Spy Streptococcus pyogenes

STGGB Skim milk tryptone glucose glycerol broth

TLR Toll-like receptor

TM Tympanic membrane

TNF Tumour necrosis factor

Treg Regulatory T

Tris Tris(hydroxymethyl)aminomethane

U Unknown

URT Upper respiratory tract

VT Ventilation tube

VTI Ventilation tube insertion

WHO World Health Organisation

WU WU polyomavirus

1

Chapter 1:

Literature review

2

1.1 Systematic review

A systematic review of otopathogens detected in the middle ear of children with

otitis media (OM) globally has been published on PloS One Journal (Appendix

A).

The published review demonstrated the bacterial aetiology of OM, including its

varied presentations and how these vary regionally and temporally since 1970. A

key finding was the persistence of the three main bacteria, Streptococcus

pneumoniae (Spn), Haemophilus influenzae (Hi) and Moraxella catarrhalis (Mcat)

detected within the middle ear effusion (MEE) of children with different types of

OM around the world. In addition, the introduction of pneumococcal conjugate

vaccines has had a demonstrated impact on the change in pneumococcal

serotypes, but also resulted in temporarily shifting the predominant otopathogen

detected in the middle ear. The review clearly evidenced that although culture

remains the gold standard for bacterial detection, PCR can improve the accuracy

and precision of identification of otopathogens in the MEE of children OM.

Importantly, the major contribution of PCR techniques was to increase

otopathogen detection rates in culture negative MEE samples, specifically, the

detection rate of Mcat was increased in otherwise “sterile” culture negative

samples. Overall, despite the observed changes, the review concluded that

continuous surveillance of otopathogens, using appropriate detection methods

will continue to provide vital information for development of appropriate treatment

and progress the development of a wider range of potential vaccines and other

novel preventive strategies.

3

1.2 Otitis media

OM is one of the most common childhood infections around the world, that, in

developed countries, also frequently results in the prescription of antibiotics by a

medical practitioner (Cripps and Kyd, 2003, Klein, 2000). This disease represents

a major health burden not only on children and their families, but also to the wider

healthcare system and society. It has been established that the estimated annual

burden of disease for OM treatment is approximately US$3-4 billion in the United

States, US$1.8 billion in Japan, US$600 million in Canada, and up to US$315

million (AU$400 million) in Australia (Coyte et al., 1999, O'Brien et al., 2009,

Taylor et al., 2009, Ahmed et al., 2014).

Conductive hearing loss, when sound is not conducted efficiently through the

outer ear canal to the eardrum and ossicles of the middle ear, is the most common

complication of OM and is the major burden of this disease. In children, hearing

loss at a young age can lead to delays in speech and language development

(O'Connor et al., 2009) but may also result in other less common but more serious

complications associated with OM including intra-temporal complications

(mastoiditis, cholesteatoma, chronic tympanic membrane (TM) perforation, facial

nerve dysfunction) and extra-temporal, intra-cranial complications (meningitis

and brain abscess) In Australia, OM is still a significant public health concern of

the country, especially in Indigenous communities (John et al., 2013, Jervis-

Bardy et al., 2014, Brennan-Jones et al., 2015). Each year, 3-5 children die due

to complications associated with OM. Approximately 15 children will suffer

permanent hearing loss, 30,000 will suffer temporary mild to moderate hearing

loss, 60,000 will suffer chronic suppurative OM (CSOM) and 90,000 will develop

a TM perforation (O'Connor et al., 2009).

OM presents clinically as a spectrum of disease and patients are primarily

diagnosed with acute otitis media (AOM) or otitis media with effusion (OME) in

urban communities (Kong and Coates, 2009) both as single episodes but

frequently as repeated or recurrent episodes.

Overall, left untreated, OM caused 3,599 deaths worldwide in 2002, due primarily

to the spread of mastoid and intracranial infection (Acuin, 2007), although

4

between one to five deaths each year in Australia (1997-2006) resulted from

suppuratives and unspecified OM (Access Economics, 2009).

1.2.1 Acute otitis media

Acute otitis media (AOM) is characterised by the rapid onset of local and systemic

symptoms, including ear pain (otalgia), fever, vomiting and development of

middle ear fluid as well as red bulging TM (Figure 1-1). Approximately 80% of

children will experience one or more AOM episodes within the first 36 months of

life (Rovers, 2008) with a peak incidence occurring in children between 6 and 18

months (Kilpi et al., 2001). Each year, more than 700 million cases of AOM are

estimated to occur globally and more than 50% of cases are reported for children

less than five years of age (Monasta et al., 2012). For most children (80%), in the

absence of antibiotic treatment, AOM usually resolves, which was defined as

resolution or improvement of AOM signs and symptoms, within 3 days (O'Neill et

al., 2007). Unfortunately, there remain approximately 15 – 20% of pre-school

children, with an average age of 15 months at diagnosis (Rosenfeld and Kay,

2003) for whom AOM does not resolve. Furthermore, up to 65% of children will

suffer recurrence of AOM (RAOM), during which children experience up to three

episodes of AOM within 6 months or more than three episodes within 12 months,

by 5 years of age (Rovers, 2008).

1.2.2 Otitis media with effusion

Otitis media with effusion (OME) is defined as the presence of middle ear fluid

(Figure 1-1) without signs or symptoms associated with acute infection or AOM

(Rovers, 2008, Bluestone, 2003). OME is one of the most common presentations

of OM and is often experienced in conjunction with rhino-sinusitis, or following an

episode of acute infection (Morris and Leach, 2009). Many episodes of OME

resolve spontaneously within 3 months (American Academy of Family

Physicians, 2004), however, unfortunately between 30 - 40% of children will

experience chronic OME (COME) whereby MEE persists for more than 3 months

(Alles et al., 2001, Bluestone, 2003). Persistence of OME for more than 4 months

may require ventilation tube insertion to aid resolution, depending on the child’s

hearing status, developmental risk, access to health care and/or risk of structural

5

damage to TM or middle ear (American Academy of Family Physicians, 2004,

Rosenfeld et al., 2016).

Figure 1-1: Normal middle ear, acute otitis media and otitis media with effusion

(Society for middle ear disease, 2012). a) Normal middle ear with clear tympanic

membrane and ossicles visible; b) AOM with red and bulging tympanic

membrane; c) OME with fluid behind the ear drum.

1.3 Pathogenesis of OM

OM is a multi-factorial disease influenced by a range of risk factors which are

shown in Figure 1-2. These factors may be broadly considered in relation to two

elements: 1) microbial carriage (microbial load, early carriage and mixed

colonisation) and 2) reduced or impaired host immune responses (Rovers, 2008,

Sun et al., 2012, Smith-Vaughan et al., 2006). The first element may be affected

by environmental factors, including number of siblings, overcrowding, poor

hygiene and attending day-care (Lehmann et al., 2008, Jacoby et al., 2011,

Gisselsson-Solén et al., 2014) (Table 1-1). The second element may vary in

response to due to host factors, including young age, male gender, genetic

predisposition and atopy (Kreiner-Møller et al., 2012, Passali et al., 2014, Engel

et al., 1999, Nokso-Koivisto et al., 2014, Allen et al., 2013, Hafrén et al., 2015)

(Table 1-1). The potential interactions between the microbial carriage and the

host immune responses strongly influence the risk of OM pathogenesis.

6

Figure 1-2: Key factors of pathogenesis of otitis media. Three key factors,

including ET dysfunction, microbial load, and immune response, and other risk

factors, including age, gender, genetic predisposition, atopy, poor hygiene,

siblings, attending day-care and overcrowding, are involved in development of

OM. Adapted from (Rovers et al., 2004).

7

Table 1-1: Risk factors for OM (Kong and Coates, 2009).

Risk Factor Comment NHMRC level of evidence*

Host-related

Age Highest incidence between 6 and 11 months

A

Sex Slightly higher preponderance among males

C

Ethnicity Indigenous children are at higher risk of earlier and more severe disease

A

Premature birth Increased risk C

Allergy Link noted but pathways unclear D

Immuno-suppression Subtle immune deficiencies often noted in RAOM

A

Genetic predisposition

Familial clustering noted A

Craniofacial abnormalities

Increased incidence in children with cleft palate, Down syndrome and craniofacial abnormalities

C

Adenoids Infected adenoids or tissue increase risk more than size of adenoids

C

Gastro-oesophageal reflux

Link noted, but further study required

D

Environmental

Day-care or overcrowding

Higher incidence with day-care attendance

B

Siblings Increased risk with older siblings B

Upper respiratory tract infection

Viruses predispose to OM B

Seasonality Increased incidence in winter months

D

Cigarette smoke exposure

Increased risk B

Breastfeeding Has a protective effect C

Socioeconomic status

Variable but generally increased risk with lower status

C

Pacifier (dummy) use Increased risk after age 11 months B

NHMRC – National Health and Medical Research Council

* NHMRC level of evidence: A – body of evidence can be trusted to guide practice; B – body of evidence can be trusted to guide practice in most situations; C - body of evidence provides some support for recommendation(s) but care should be taken in its application; D – body of evidence is weak and recommendation must be applied with caution.

8

1.4 Microbiology of OM

The role of microbial exposure in the development of OM is complex, since

multiple microbial agents, bacterial and viral may be normally present in the upper

respiratory tract, but may also act either independently or together to cause OM

pathogenesis (Massa et al., 2009).

1.4.1 Role of bacteria in OM

1.4.1.1 Three main otopathogens

Bacterial infections of the upper respiratory tract (URT), which involves the

mouth, nose, and pharynx superiorly and the glottis inferiorly (Perkin et al., 2007),

are associated with 88% of AOM and 84% of OME episodes (Bluestone, 2003)

with common commensal bacteria commonly involved. The nasopharynx and

adenoids are potential sources of the predominant bacteria associated with OM

pathogenesis. Globally, the three predominant bacterial otopathogens associated

with OM pathogenesis are Streptococcus pneumoniae, (Spn), non-typeable

Haemophilus influenzae (NTHi), and Mcat (Ngo et al., 2016).

For most countries around the world, Spn was the predominant pathogen isolated

from the middle ear and nasopharynx of children with AOM (Casey et al., 2010,

Dupont et al., 2010), prior to the introduction of the heptavalent pneumococcal

conjugate vaccine (PCV7) which included 7 common serotypes (4, 6B, 9V, 14,

18C, 19F and 23F). Following the widespread of implementation of PCV7, there

has been gradual serotype replacement of the pneumococcal serotypes isolated

from children with OM, to non-PCV7 serotypes, especially 19A (Casey et al.,

2010, Dupont et al., 2010, Hoberman et al., 2011, Alonso et al., 2013, Fenoll et

al., 2011, Gene et al., 2004). Despite ongoing vaccine development and

implementation, Spn remains the predominant bacterial otopathogen for AOM but

not COME in most countries (Ngo et al., 2016).

The ten-valent pneumococcal NTHi protein D conjugate vaccine (PHiD-CV),

which concludes additional serotype 1, 5 and 7F, and protein D of NTHi, was

introduced into NIP in 2010 some countries, such as Brazil and Finland (Jokinen

et al., 2015, Moreira et al., 2016) and was licensed for use in other countries,

9

such as Australia, Malaysia and Singapore and Japan [Leach et al., 2015; Lim et

al., 2014]. The vaccine showed benefit in reduction middle ear infection due to

NTHi but Spn in Australian Indigenous children from remote communities [Leach

et al., 2015]. PHiD-CV also showed more effective than the 13-valent

pneumococcal conjugate vaccine (PCV13) in prevention of AOM development in

children (Shiragami et al., 2015).

Recently, vaccination with PCV13, which concludes serotype 1, 3, 4, 5, 6A, 6B,

7F, 9V, 14, 18C, 19F, 23F and 19A, has replaced PCV7 in many developed

countries. A similar scenario of pneumococcal serotype replacement has been

observed in recent studies, which confirmed a reduced prevalence of the 19A

serotype isolated in the middle ear samples after PCV13 introduction in the US

(Kaplan et al., 2015) and Israel (Ben-Shimol et al., 2014). These studies also

revealed the emergence of non-PCV13 pneumococcal serotypes, especially

serotypes 15 and 35B, isolated from children with OM (Ben-Shimol et al., 2014,

Kaplan et al., 2015, Ozawa et al., 2015, Martin et al., 2014b, Angoulvant et al.,

2015).

Additionally, since the widespread introduction of pneumococcal conjugate

vaccines, a change to the predominant otopathogens identified from the middle

ear has been observed. In several countries, including the USA, NTHi became

prevalent in children with AOM immediately after PVC7 introduction (Casey et al.,

2010, Dupont et al., 2010, Grubb and Spaugh, 2010, Murphy, 2015, Tamir et al.,

2015, Dagan et al., 2016). Interestingly, Mcat was recently reported as the most

frequent bacteria causing RAOM in one study (Casey et al., 2015b) and the

second most common bacteria detected in the middle ear of children with OM in

two additional studies (Mills et al., 2015, Holder et al., 2015), all of which

examined children after PCV13 implementation.

NTHi is a bacterium with a strong association with chronic and recurrent AOM

(Barkai et al., 2009). The standard laboratory culture systems do not readily

distinguish between this bacteria and its close relative, Haemophilus

haemolyticus, a human respiratory tract commensal (Murphy et al., 2007), which

may, in the past, have resulted to over representation through misidentification of

NTHi (Kirkham et al., 2010). Increased use of real-time PCR assays, which target

10

different genome regions of NTHi, such as protein D gene (Binks et al., 2012) or

L-glucose permease coding locus (Price et al., 2015) have improved detection

frequencies and the capacity to distinguish between the 2 organisms.

The third most prevalent bacterium associated with OM pathogenesis is

Moraxella catarrhalis (Mcat) (Murphy and Parameswaran, 2009, Murphy, 2005,

Balder et al., 2013). AOM involving Mcat infections are more often associated

with first episodes of AOM, in children at a younger age of OM onset (Broides et

al., 2009). Virtually all Mcat isolates from MEE of children with OM are β-

lactamase positive. This characteristic results in increasing their resistance to

penicillin-based antibiotics, which are typically the front line medications

prescribed to treat OM (Wiertsema et al., 2011b, Dupont et al., 2010). Alarmingly,

Mcat has been reported to have overtaken Spn and NTHi as the most common

bacterial cause of RAOM in the US children (Casey et al., 2015b) in era of PCV13.

1.4.1.2 The origin of bacterial otopathogens

The nasopharynx and adenoids are recognised as potential within host locations

that may act as a reservoir to seed viruses or bacterial otopathogens into the

middle ear (Massa et al. review). Early and heavy colonisation of nasopharynx

with these bacteria has been closely correlated with the early onset of OM.

Australian Indigenous children, who are classified a population at high risk of OM

(Gunasekera et al., 2007a, Morris et al., 2007, Watson et al., 2006), have been

reported as being colonised by these bacteria by as early as 21 days of age

(Smith-Vaughan et al., 2008) and up to 95% of these children are diagnosed with

either AOM or OME by between 6-8 weeks of age (Boswell and Nienhuys, 1996,

Boswell and Nienhuys, 1995). In contrast, non-Indigenous Australian children do

not typically experience the same load and frequency of colonisation by these

bacteria within the first year of life (Leach et al., 1994) with approximately 30%

of children diagnosed with OME (Boswell and Nienhuys, 1996). Previous reports

have provided evidence of nasopharyngeal colonisation, bacterial load and

correlation with OM diagnosis for Indigenous Australian children, however less is

known regarding these parameters for urban Australian children (Marsh et al.,

2012, Binks et al., 2011).

11

The association between bacterial colonisation of the URT and otitis media

identified in Australian Indigenous children may result from seeding of bacteria

from other URT locations. Colonisation of the adenoid tissues has been

associated with OM pathogenesis, primarily through transmission of the

otopathogens via the Eustachian tubes (ET) into the middle ear cavity (Gates,

1999). This hypothesis is supported by more frequent detection of the three main

otopathogens in the adenoids of children with OM compared to that observed

from children not experiencing OM (Tomonaga et al., 1989, Suzuki et al., 1999,

Karlidağ et al., 2002). Furthermore, there is genetic similarity between the specific

otopathogenic strains isolated from adenoids and MEE within the same children,

diagnosed with COME (Emaneini et al., 2013). Further evidence, such as

comparisons of the bacteria colonising multiple URT locations within the same

child is currently limited, although would contribute to our understanding of the

origin of middle ear infections.

1.4.2 Role of viruses in OM

Viruses induce inflammation of the nasopharynx by stimulation of local release of

cytokines and inflammatory mediators, ultimately increasing the risk of OM

pathogenesis, by increasing the opportunity for bacterial adherence and

colonisation. Together, these actions increase the risk of ET dysfunction.

Furthermore, viral infections have the capacity to suppress or disrupt host

immune responses, which can also contribute significantly to the susceptibility to

AOM pathogenesis (Bakaletz et al., 1993, Heikkinen, 2000, Heikkinen and

Chonmaitree, 2003, Chonmaitree et al., 2008b).

Infection of the URT by live viruses may initiate or induce the inflammatory

processes that contribute to subsequent infection of the middle ear by bacteria,

although viruses alone, in the absence of bacteria, can also cause AOM

(Chonmaitree et al., 2012). Viruses have been detected in 70% of middle ear

fluids from patients with AOM, with 4% of cases identified as viruses alone

(without bacteria detected) (Chonmaitree et al., 2012).

Common respiratory viruses are frequently associated and identified within

children diagnosed with AOM and these include adenovirus (17-46% of OM

12

cases), respiratory syncytial virus (41-56% of OM cases), rhinovirus (30-44% of

OM cases), influenza virus (23-35% of OM cases) and coronavirus (50% of OM

cases) (Bulut et al., 2007, Yano et al., 2009a, Wiertsema et al., 2011a).

Co-infection of the middle ear by both bacteria and viruses have been reported

in 15% of MEE samples collected from children with AOM (Chonmaitree, 2000)

(Table 1-2). The actual frequency of viral detection within previous studies could

be under-reporting actual incidence since improved methodologies such as

polymerase chain reaction (PCR) detect viruses more efficiently than state of the

art viral culture/viral antigen detection methods. For example, the frequency of

co-infection by virus and bacteria in MEE for children experiencing OM was 66%

higher when reported using PCR techniques (Ruohola et al., 2006) (Table 1-2).

Increasing use of PCR for viral detection in more recent reports reinforce the need

to consider detection methods when comparing viral detections from earlier

reports.

Table 1-2: Bacterial and viral findings of MEE from children with AOM

(Chonmaitree et al., 2012).

Bacterial-viral culture and RSV-EIA*

Conventional and molecular techniques

Viruses alone 5% 4%

Bacteria alone 55% 27%

Viruses and bacteria 15% 66%

No pathogen 25% 4%

* RSV-EIV: respiratory syncytial virus was detected by enzyme immunoassays

1.4.3 Biofilms in OM

Biofilms are defined as complex organisations of bacteria embedded in a slime-

like, extracellular polymeric matrix consisting of polysaccharides, nucleic acids

and proteins (Figure 1-3) (Costerton et al., 2003, Hall-Stoodley and Stoodley,

2002). Bacteria within biofilms are difficult to culture, resistant to antibiotics and

host defence mechanisms (Ehrlich et al., 2002) and thus are under increasing

13

examination for their important role in pathogenesis of chronic diseases such as

chronic or recurrent OM.

Otopathogenic biofilms have previously been demonstrated in 92% of middle ear

biopsies of children suffering RAOM and COME (Hall-Stoodley et al., 2006).

These biofilms were observed in 65% of middle ear mucosal biopsies from a large

cohort of children with RAOM and COME undertaken in Western Australia

(Thornton et al., 2011). Examination of biofilms using fluorescent in situ

hybridisation (FISH) and confocal laser scanning microscopy (CLSM)

demonstrated multiple bacterial species within the structures (Post et al., 2007).

The frequent presence of biofilms within the middle ear contributes to our

understanding of the frequent ineffectiveness of current treatment strategies,

particularly antibiotic treatment failure, aimed at preventing and/or resolving

RAOM and COME. The specific role of biofilms in disease pathogenesis for OM

requires further research to better define opportunities to prevent or deconstruct

them (Thornton et al., 2011).

Middle ear mucosal samples are challenging to obtain and so there has been

recent interest in the use of middle ear effusate (MEE) aspirated from the middle

ear. Biofilms have been identified and reported in the MEE of children with RAOM

using FISH and CSLM (Thornton et al., 2013). Larger scale monitoring of the

effectiveness of clinical interventions to treat biofilms and the resulting OM may

potentially utilise MEE sampling since it may be equally efficient and less

intervention than the use of middle ear biopsies (Thornton et al., 2013).

The middle ear is not the only URT location in which biofilms have been identified.

Biofilms have been demonstrated on the adenoids of otitis prone children (Hoa

et al., 2009, Kania et al., 2008, Nistico et al., 2011) and these biofilms also

contained the three predominant otopathogens, causal for OM including Spn,

NTHi and Mcat (Hoa et al., 2009, Nistico et al., 2011). These results support the

hypothesis that adenoid biofilms may contribute as a potential reservoir for

otopathogens that facilitate reinfection of the middle ear (Hoa et al., 2009, Nistico

et al., 2011).

14

Figure 1-3: The general process of bacterial biofilm formation (Stoodley and

Dirckx, 2003). 1) Planktonic bacteria are attaching to surface and start producing

extracellular polymeric substance; 2) Self-organised and highly-structure 3D

biofilm community appear; 3) Intercellular interaction and signalling within the

community and environment. Clumps of cells are released to colonise new

surfaces.

1.5 Immunology of OM

Local and systemic immune responses within the host are important elements in

the prevention, protection and response mucosal infection of OM. The airway

epithelium acts as the first line of defense against invasion of microbes and

utilises a wide range of defence mechanisms, including mechanical, such as

mucus, innate and adaptive elements (Massa et al., 2015). When bacteria and

viruses interact with airway epithelial cells, antimicrobial agents and chemical

signalling agents are induced as part of the innate immune response and

adaptive immune system is engaged (Ogra, 2010, Vareille et al., 2011).

1.5.1 Cytokines in OM

Cytokines play critical roles as initiators, mediators and regulators of the middle

ear inflammation, leading to histopathological changes in the middle ear cavity

and pathogenesis of OM (Smirnova et al., 2002).

15

The early inflammatory mediators, which initiate acute inflammatory reactions

against infections, have been identified in MEE of children with a variety of

presentations of OM. Cytokines such as tumour necrosis factor-alpha (TNF-α)

were identified from children with COME (Ophir et al., 1988, Maxwell et al., 1997,

Skotnicka and Hassmann, 2000), interleukin 1-beta (IL-1β) from children with

AOM (Harimaya et al., 2009) and COME (Juhn et al., 1992, Juhn et al., 1994).

Interleukin 8 (IL-8) has also been identified in children with COME (Takeuchi et

al., 1994, Hotomi et al., 1994, Sekiyama et al., 2011) and AOM (Storgaard et al.,

1997, Leibovitz et al., 2000).

Cytokines that operate as ongoing inflammatory mediators, such as interferon

gamma (IFN-γ) and interleukin 6 (IL-6), have been detected in MEE of children

with OM (Yellon et al., 1992, Yellon et al., 1995, Pitkäranta et al., 1996).

Furthermore, immunoregulatory cytokines, such as interleukin 5 (IL-5) and

interleukin 10 (IL-10) have also been identified in the MEE of children with OM

(Wright et al., 2000, Skotnicka and Hassmann, 2000). Thus, whilst a wide range

of cytokines have been identified in serum and MEE of children with AOM against

acute infections (Chkhaidze et al., 2007, Patel et al., 2009, Liu et al., 2013b),

however, there is a relative paucity of information regarding the relationship

between local and systemic cytokine responses in children with RAOM or COME.

The roles of these cytokines in RAOM and COME are still unclear.

Individual cytokine gene polymorphisms relating to OM susceptibility and severity

were described concisely in a recent comprehensive review (Mittal et al., 2014).

For example, TNF-α(-308), IL-6(-174), IL-10(-592), IL-10(-819) and IL-10(-1082)

polymorphisms were reported with respect to increasing risk for OM susceptibility

and replacement of ventilation tubes (Patel et al., 2006, Emonts et al., 2007,

Revai et al., 2009, Ilia et al., 2014, Nokso-Koivisto et al., 2014), whilst IL-

1β(+3953) was associated with the risk of more severe AOM (McCormick et al.,

2011).

1.5.2 Serum antibody responses in OM

OM pathogenesis is frequently caused by Spn and Hi and a number of studies

have reported on the presence of host antibody responses developed against

16

antigens of Spn and Hi in children with OM. The variability of serum antibody

responses developed against pneumococcal antigens has proven challenging to

interpret with some reports suggesting that children with OM, particularly AOM

have reduced levels of antibody response, whilst others remain the same or

increased. For example, children with RAOM are reported to have lower levels of

anti- pneumococcal serotype or anti- polysaccharides IgG, and IgA (Freijd et al.,

1984, Veenhoven et al., 2004, Prellner et al., 1989, Dhooge et al., 2002).

Veenhoven et al. (2004) also reported that children with recurrent episodes of

AOM had lower level of serum IgG2 against pneumococcal polysaccharides

compared with values from children without RAOM (Veenhoven et al., 2004).

Another study revealed that otitis prone children had significantly lower levels of

serum IgA and IgG2 against the 7 pneumococcal serotypes, included in PCV7,

compared to a control population. This result may suggest that otitis prone

children fail to respond in these subclasses after vaccination with pneumococcal

conjugate vaccine (Dhooge et al., 2002). A more recent study has reported that

children identified as otitis-prone and AOM treatment failure, mount less of a

serum IgG response than children non-otitis prone to Spn proteins (PhtD, LytB,

PhtE and Ply) following AOM (Kaur et al., 2011a). The results of poor serum IgG

response in otitis-prone children may be explained by poor memory-B cell

generation in these children (Sharma et al., 2012).

Clearly the relationship between systemic antibody responses and OM risk is not

a simple one, since a number of reports indicate that antibody responses to

pneumococcal antigens maybe unchanged or increased in children with OM. For

example, reports of either normal or higher levels of IgG, IgA, and its subclasses

against pneumococcal polysaccharides in children with OM (Berman et al., 1992,

Drake-Lee et al., 2003, Misbah et al., 1997, Sørensen and Nielsen, 1988).

Children with recurrent OM showed normal serum antibodies against

pneumococcal polysaccharides, compared to healthy control (Berman et al.,

1992) whilst another study revealed that children with recurrent OM had

significant higher levels of serum antibodies against pneumococcal

polysaccharides, compared to those without recurrent OM (Misbah et al., 1997).

Similarly, children with RAOM showed normal or higher levels of serum IgG and

its subclasses against pneumococcal proteins (Wiertsema et al., 2012) and

17

polysaccharides (Menon et al., 2012, Corscadden et al., 2013, van Kempen et

al., 2006).

This pattern of conflicting results is also seen in serum antibodies to antigens of

Hi in children with OM. Otitis prone children have been shown to mount a lower

serum IgG response against Hi proteins after AOM, which may account for

recurrent infection (Pichichero et al., 2010, Kaur et al., 2011b). In contrast, IgG

responses to Hi proteins (P4, P6 and PD) were not impaired in children with

RAOM (Wiertsema et al., 2012).

Serum antibody response to 5 proteins of Mcat has been reported in only one

study which examined antibody responses after nasopharyngeal colonisation in

children with AOM (Ren et al., 2015). That study reported that although serum

antibody levels did not differ between children who did or did not progress to AOM

after nasopharyngeal colonisation, the high antibody levels against three proteins

(OppA, Msp22 and Hag) were correlated to reduction of Mcat carriage (Ren et

al., 2015).

1.5.3 Mucosal antibody responses in OM

OM may be considered a mucosal infection, thus mucosal antibodies, especially

secretory IgA (sIgA), could act to minimise bacterial adherence and improve

bacterial clearance through opsonisation and phagocytosis locally within the

middle ear (Massa et al., 2015). Pathogen specific antibodies from the middle ear

and saliva have been measured in children with OM and have identified

antibodies against pneumococcal polysaccharides and proteins in more than

50% of MEE and salivary samples (Simell et al., 2007, Virolainen et al., 1995,

Howie et al., 1973, Verhaegh et al., 2012) and are associated with clearance of

the infection (Simell et al., 2007, Karjalainen et al., 1991a, Karjalainen et al.,

1991b). It remains unclear whether these antibodies are produced locally or

systemically, however IgA secretion is generally considered to be derived from

both local and systemic production, whilst IgG is predominantly derived

systemically from blood.

18

1.5.4 Regulatory T cells in OM

Regulatory T cells (Treg cells) have a crucial role in prevention of autoimmunity

and maintenance of immune homeostasis (Delgoffe et al., 2013) by regulation of

the host immune response but may also contribute to recurrence and chronicity

of infection (Zhang et al., 2011). In OM pathogenesis, the proportion of blood

derived Treg cells (CD4+CD25+FoxP3+ T cells) identified within adults with

COME was significantly higher than that observed within patients diagnosed with

acute OME or healthy controls (Zhao et al., 2009). Furthermore, similar

percentages of CD4+CD25+ T cells and CD8+CD25+ T cells were observed in

the blood and adenoids of children with OME and adenoid hypertrophy (AH) ,

compared to children with AH only (Kotowski et al., 2011). In general, the role/s

of Treg cells in regulation of host immune responses to otopathogens and their

impact on the recurrence and chronicity of OM in children requires further

investigation to improve our understanding and guide strategies for treatment and

prevention of OM.

1.6 Systemic gene expression in OM

Although OM is a multi-factorial disease of the middle ear cavity, there are key

events within OM pathogenesis, specifically the interrelated roles of microbial

infection and host immune responses including inflammatory responses against

the infection that contribute significantly to pathogenesis of OM within individuals.

Host immune responses to environmental and infective agents that may increase

the risk of OM, include innate and adaptive as well as local and systemic

responses. Increasingly, the role of gene expression that influences or regulates

these immunity networks and their signalling are being explored as opportunities

to either identify individuals at higher risk of OM or to facilitate development of

alternate approaches for OM prevention and treatment.

Systemic gene expression during onset of AOM has been investigated using

animal models and in humans models. In animal models, significant gene-based

regulation of a robust inflammatory response was demonstrated during Spn

infection (Chen et al., 2005). In contrast, positive and negative regulation of

immune gene expression were observed during AOM in response to Hi infection

19

(Hernandez et al., 2015). Together, these reports emphasise the importance of

gene expression in the regulation of anti-inflammatory responses in influencing

OM pathogenesis. In humans, significant up-regulation of genes associated with

bacterial defences and down-regulation of genes associated with cell-mediated

immune responses were observed during Spn infection causal for AOM (Liu et

al., 2012). In contrast, during Hi induced AOM, a majority of genes associated

with a pro-inflammatory cytokine response were downregulated (Liu et al.,

2013a). Together these studies highlight the need to recognise the role of the

specific pathogens and the differential regulation of genes associated with host

immune responses during AOM.

Interestingly, increased level expression of a single gene, such as HLA-A2, or

mutation in specific genes, such as Fbxo11, which is expressed in middle ear

epithelial cells and encodes of a regulator of TGF-β signalling pathway, have

been reported from patients experiencing RAOM/COME and was associated with

the chronicity of OM (Kalm et al., 1994, Kalm et al., 1991, Rye et al., 2011). In

addition, mutations in genes relating to primary ciliary dyskinesia, such as

CCDC114, DNAI1, and DAAF2, may be associated with OM (Mata et al., 2014).

Further investigation is needed to investigate and compare differences in

expression of a wide range of genes between children diagnosed with

chronic/recurrent OM and either healthy children or children with active AOM

infections.

20

1.7 OM treatment

1.7.1 Antibiotics

Treatment for AOM is a predominant reason for young children around the

developed world, to be prescribed antibiotics (Grijalva et al., 2009). Antibiotics

are frequently prescribed to treat the bacterial aetiology of the disease in general

practice (Heikkinen and Chonmaitree, 2003, Lieberthal et al., 2013). Microbial

pathogenesis of OM may however, include viruses alone and/or together with

bacterial (Heikkinen et al., 1999, Heikkinen and Chonmaitree, 2003). A meta-

analysis of OM treatment has shown that the effectiveness of antibiotic treatment

of OM varies, with antibiotics providing a 14% increase in clinical improvement of

the symptoms experienced during AOM compared to placebo or no drug controls

(Rosenfeld et al., 1994). Two recent studies, involving randomised trials using

stringent diagnostic criteria for AOM in young children (<36 months of age),

showed that antibiotic therapy may increase clinical improvement from 26% to

35% as compared with placebo (Hoberman et al., 2011, Tähtinen et al., 2011).

Greater benefits for patient recovery were observed when immediate antibiotic

therapy was instituted for the treatment of bilateral AOM (Rovers et al., 2006,

McCormick et al., 2007), or AOM associated with ear discharge, pneumococcal

infection (Rovers et al., 2006) and in children younger than 2 years of age

(Gunasekera et al., 2007b, Hoberman et al., 2011). In contrast, more than 85%

of AOM experienced by children will resolve spontaneously within 2 weeks of

onset with either no treatment or placebo treatment (Rosenfeld et al., 1994, Del

Mar et al., 1997, Damoiseaux et al., 1998, Gunasekera et al., 2009). Antibiotic

therapy does not immediately resolve acute symptoms of AOM, such as pain

and persistence of MEE (Sanders et al., 2009, Koopman et al., 2008) and may

result in increased risk of side effects such as diarrhoea, vomiting and eczema

in children (Sanders et al., 2009, Tähtinen et al., 2011).

The efficacy of different antibiotic types on AOM resolution is difficult to monitor

due to constantly changing bacterial strains and serotypes. Meanwhile the

development of increased bacterial resistance to commonly prescribed antibiotics

remains a major public health challenge. In developed countries, there is

increased scrutiny of the use of antibiotics for the treatment of children at low risk

21

of development of severe complications for OM reports on the rate of treatment

failure, defined as children who were prescribed with different classes of

antibiotics within 2-18 days after the first prescription, of different classes of

antibiotics on AOM (McGrath et al., 2013) demonstrated a range of antibiotic

failure rates. Macrolides had the lowest rate of failures in AOM treatment in

children (below 9%), while other antibiotic classes, such as cephalosporins,

amoxicillin, unspecified penicillin, and amoxicillin/clavulanate, had failure rates

around 10% (McGrath et al., 2013). High dose Amoxicillin (80–90 mg/ kg per day

in 2 divided doses) is recommended as the first line treatment in the absence of

a known allergy and cephalosporins should be alternatively used in children who

are allergic to penicillin (Lieberthal et al., 2013, Harmes et al., 2013). Amoxicillin

use has a number of advantages, including low cost, effectiveness and a narrow

microbiologic spectrum, which could reduce antibiotic resistance rate

development within susceptible bacteria (Harmes et al., 2013). In the case of

antibiotic treatment after 48-72h of failure of the initial antibiotic treatment,

intramuscular or intravenous ceftriaxone should be used effectively (Harmes et

al., 2013, Lieberthal et al., 2013). It is likely, however, that misuse of this agent

may result in increased high-level penicillin resistance within the community

(Harmes et al., 2013).

In OME management, antibiotics are considered to perform a limited role in OME

treatment (Rosenfeld and Kay, 2003) and do not hasten the clearance of middle

ear fluid (American Academy of Family Physicians, 2004, Harmes et al., 2013).

Long-term administration of antibiotics, continued for between four weeks to three

months, until OME is completely resolved, has been reported to reduce the

incidence of glue ear (van Zon et al., 2012). The benefits of long term antibiotic

treatment for OME is generally considered not sufficient to prescribe them to

children at low risk of severe complications of OM. The spontaneous resolution

rate of OME, whereby more than 70% of children will resolve clearance of the

middle ear cavity naturally within 3 months (Rosenfeld et al., 2004, Rosenfeld

and Kay, 2003), further supports minimal use of antibiotics in OME management

(Harmes et al., 2013).

22

In Australia, antibiotics are recommended for AOM children aged younger than 6

months and all Indigenous children with AOM (Gunasekera et al., 2007b,

Gunasekera et al., 2009). Indigenous children in Australia are classified as a

group potentially at high risk of development of OM, with approximately 40% of

children (aged 3–30 months) suffering OME, 25% suffering AOM without

perforation (AOMwop) and 5% suffering AOM with perforation (AOMwp) (Morris

et al., 2007). Children at high risk of OM maybe prescribed prophylactic antibiotic

therapy. Long term (24 weeks) antibiotic therapy was prescribed for Indigenous

children with OM (Leach et al., 2008) and although there were no differences

observed in rates of children with OME or AOMwop, amoxicillin significantly

reduced the risk of perforation (Leach et al., 2008). Furthermore, administration

of a single dose of azithromycin has been shown to reduce nasal carriage of Spn

and Hi in Indigenous children (aged 6 months–6 years), although the rate of

clinical failure was high (Morris et al., 2010). Overall, prophylactic use of

antibiotics in the longer term may provide some benefit to children at high risk of

severe and frequent OM infection, through reduction of bacterial carriage.

1.7.2 Surgery

Children with RAOM and COME may be considered for ventilation

(tympanostomy) tube insertion (VTI) commonly referred to as grommets, and/or

adenoidectomy. VTI is one of the most common operations in children from

developed countries (Lous et al., 2005, Kong and Coates, 2009, Rosenfeld et al.,

2013, Rosenfeld et al., 2016) and is used to improve ventilation and pressure

regulation in the middle ear (Lous et al., 2005). VTI are used to treat children

aged from 6 months to 12 years who have a significant history of bilateral COME

or RAOM with MEE and/or hearing loss at the time of assessment (Rosenfeld et

al., 2013, Rosenfeld et al., 2016, Harmes et al., 2013). Children treated using VTI

for OME, have been reported to experience 32% shorter period of effusion within

the first year compared to children permitted to resolve the effusion naturally

(Lous et al., 2005). Hearing deficits arising from OM VTI for children with OME,

improved their hearing levels by 6 to 9 dB after 6 to 12 months (Lous et al., 2005).

Characteristically, for children at ongoing risk of OM, up to one third of children

may require reinsertion of ventilation tubes (Coyte et al., 2001). VTI has been

23

associated with otorrhae, with rates ranging from 26% to 83%, with approximately

24% of otorrhae cases following VTI involving S. aureus infection (Marzouk et al.,

2011). Methicillin-resistant S. aureus (MRSA) has also been identified in

ventilation tube-related otorrhoeae in paediatric patients (Cheng and Javia, 2012).

In addition, VTI is also associated with TM perforation in 13%-18% of children

after the surgery (Lous et al., 2011, Hong et al., 2014).

Adenoidectomy is another frequent childhood surgical procedure, especially in

Western countries. Adenoidectomy is performed to reduce airway obstruction

and reduce the opportunity for reinfection from the adenoid tissue which may act

as an otopathogen reservoir (van den Aardweg et al., 2010). Currently, the

effectiveness of adenoidectomy alone in children with OM is unclear (van den

Aardweg et al., 2010). Adenoidectomy in combination with a unilateral VTI has

been reported to provide a positive effect on OME resolution more than 20% for

the other ear over at least 6 months (van den Aardweg et al., 2010). Combination

adenoidectomy and VTI surgery has been reported to reduce hearing loss by 5

dB as compared to VTI only (Lous et al., 2005, van den Aardweg et al., 2010). In

addition, a large population based study conducted in Australia reported that

adenoidectomy in combination with the first or subsequent VTI reduced the risk

of further VTI by 57% when compared to VTI only (Kadhim et al., 2007).

Interestingly, nasopharyngeal carriage of Spn has been reported to increase

following adenoidectomy and VTI which may suggest a possible role of adenoid

tissues in the regulation of nasopharyngeal ecology and development of systemic

immunity against Spn (Mattila et al., 2012, Mattila et al., 2010).

Overall, with regards to surgical treatment for OM, VTI has shown ongoing benefit

for children with COME or RAOM. VTI and its role in the management and

treatment of OM, with or without adenoidectomy, provides a key treatment option

for recurrent OM in developed countries.

1.8 Vaccines and OM

OM is a polymicrobial disease and thus to optimise the potential effectiveness of

vaccines in OM prevention, antigens from a range of microbes, including Spn,

NTHi, Mcat and viruses such as RSV, HRV and other respiratory viruses should

24

be considered (Cripps and Otczyk, 2006, Nokso-Koivisto et al., 2015). Although

currently unlikely that a single vaccine could protect against all pathogens that

cause OM is not feasible, a vaccine formulation that combines the predominant

otopathogens should be considered (Cripps et al., 2005, Cripps and Otczyk,

2006). There are a number of challenges in the development of a vaccine for OM.

Firstly, identification of specific antigens capable of generation of the host’s

protective immune responses, especially in young children, without causing

damage to the middle ear. Secondly, since almost all otopathogens are

commensal within the nasopharynx, a vaccine against these microbes may

negatively influence colonisation, which could result in establishment of an

ecological niche for other pathogens to colonise the nasopharynx (Cripps and

Otczyk, 2006, Devine et al., 2015, Biesbroek et al., 2014, Spijkerman et al.,

2012).

1.8.1 Bacterial vaccines

Currently, conjugate vaccines for Spn, developed for invasive pneumococcal

disease, are available for National Immunisation Programme (NIP) in many

countries. The efficacy of these vaccines in OM prevention has also been

assessed and the vaccines clearly reduce pneumococcal carriage in the

nasopharynx (Weil-Olivier et al., 2012) and increase AOM prevention (Weil-

Olivier et al., 2012). PCV7 is reported to reduce the occurrence of AOM/RAOM

in children by 6-16% (Black et al., 2000, Eskola et al., 2001). The reduction in

AOM/RAOM may not be consistent for children at high risk of OM since PCV-7

vaccination did not significantly reduce the rates of AOM and tympanic membrane

perforation in Australian Aboriginal (Mackenzie et al., 2009). Subsequent to the

inclusion of PCV7 into NIP, there has been serotype replacement identified in the

Spn strains isolated from children with OM. Non-vaccine serotypes, especially

19A, have gradually replaced vaccine serotypes (Casey et al., 2010, Dupont et

al., 2010, Pumarola et al., 2013).

A shift in Spn serotypes causal for AOM and the emergence of NTHi as the

predominant bacterial otopathogen immediately following the introduction of

PCV7, although this shift altered with the emergence of non-vaccine serotype

AOM (Casey et al., 2010, Dupont et al., 2010). Subsequently, introduction of

25

PCV13 (Leach et al., 2015) and PHiD-CV have been introduced into the NIP of

many countries, replacing PCV7 (Nuorti and Whitney, 2010, Tyo et al., 2011,

Jefferies et al., 2011, Newall et al., 2011, Lagos et al., 2011, Dicko et al., 2013,

Kim et al., 2011, Ben-Shimol et al., 2014, Weiss et al., 2015, Angoulvant et al.,

2015, Cohen et al., 2015, Leach et al., 2015). Further shifts in the predominant

bacteria and serotypes have occurred. PCV13 was reported to decrease OM a

further 19% in children younger than 10 years-old after the significant reduction

(22%) resulting from PCV7 immunisation in the United Kingdom [Lau et al., 2014].

During AOM, pneumococcal carriage decreased significantly in French children

aged 6 – 24 months after PCV13 introduction in France, especially strain 19A

and penicillin non-susceptible strains, [Cohen et al., 2015]. It is expected that

incidence of pneumococcal AOM will reduce in post-PCV13 era (Cohen et al.,

2015). While effective for pneumococcal OM, the PHiD-CV vaccine may be less

broadly protective for NTHi OM than currently envisaged from an earlier prototype

vaccine, PHiD-CV11 (van de Vooren et al., 2014, Prymula et al., 2006, Tregnaghi

et al., 2014, Vesikari et al., 2016). A recent study of Australian Indigenous

children from the Northern Territory reported that the frequency of middle ear

infection resulting from NTHi was reduced but pneumococcal OM was not

reduced in these children from remote communities after immunisation with

PHiD-CV (Leach et al., 2015).

A vaccine against Mcat, is not available although potential candidates for

development of an effective vaccine for this bacterium are being investigated

(Mawas et al., 2009, Ruckdeschel et al., 2009, Ren et al., 2015).

1.8.2 Viral vaccines

Development of viral vaccines for respiratory viruses such as RSV and

parainfluenza virus is important in prevention of viral respiratory tract infection,

including middle ear infection. Influenza vaccines have been licensed for use in

children (>6 months of age) for several decades. These vaccines are a

preparation of inactivated virus or viral subunits and administrated

intramuscularly (Cripps and Otczyk, 2006) and are reported to significantly

reduce the incidence of OM in children (aged 6 months – 5 years) (Heikkinen et

al., 1991, Clements et al., 1995, Ozgur et al., 2006, Marchisio et al., 2009). Other

26

influenza vaccines that use live attenuated virus are administrated intranasally,

and demonstrated significant reduction of OM incidence in children. Currently,

these vaccines are only recommended for children older than 2 years of age for

safety (McCarthy and Kockler, 2004, Block et al., 2011, Heikkinen et al., 2013).

In contrast to the previous studies, a number of studies have reported no change

in the incidence of OM between influenza vaccinated and placebo patients

(Hoberman et al., 2003, Jefferson et al., 2005). These studies have substantial

limitations, including a lack of an influenza epidemic during the second season

(Hoberman et al., 2003) or insufficient statistical power (Jefferson et al., 2005). In

general, although influenza vaccines have exhibited protective benefits for OM, it

is unlikely that influenza will be included in a multiple-microbial otitis vaccine due

to the almost annually changing dominant influenza strain for inclusion in the

vaccine (Cripps and Otczyk, 2006, Norhayati et al., 2015).

Vaccines for other important viruses, such as RSV and PIV, are currently not

available, although ongoing vaccine development against these viruses is a high

priority for the National Institute of Allergy and Infectious Disease, USA (Cripps

and Otczyk, 2006). A vaccine for RSV has been in phase III trial in the United

States, in which the vaccine was given to pregnant women who may transfer their

antibodies and protect their infants from the viral infection (Phillips and Flynn,

2015).

27

1.9 Significance and aims of the project

OM is one of the most common childhood infections, which frequently results in

a child visiting their general practitioner and being prescribed antibiotics (Cripps

and Kyd, 2003). OM is a multi-factorial disease with many risk factors including

young age, gender, lack of breast feeding, day-care attendance, siblings,

overcrowding, sub-optimal hygiene, low socioeconomic status, nasopharyngeal

colonisation, upper respiratory tract infection, and genetic disorders (Rovers et

al., 2004, Kong and Coates, 2009, Massa et al., 2009).

Many of these key risk factors for OM relate to two characteristics: 1) increased

microbial load due to environmental factors, including siblings, overcrowding,

day-care attendance; and 2) reduced or impaired host immune responses due to

host factors, including young age, genetic predisposition and atopy (Rovers,

2008). Thus, interactions between bacteria/viruses and the effectiveness of host

immune responses are likely to continue to play important roles in pathogenesis

of OM.

Host immune responses to microbial otopathogens within the upper respiratory

tract and middle ear have been investigated extensively, particularly over the last

thirty years, in an effort to improve the opportunity for development of

preventative vaccines (Massa et al., 2015). These studies clearly illustrate the

complex interaction between a significant variety of otopathogens, intracellular

signalling cascades and regulation of local and systemic immune responses to

the otopathogens (Massa et al., 2015). To date, comprehensive identification of

otopathogens within a number of regions of the upper respiratory tract including

the middle ear, the nasopharynx and adenoids and characterisation of the host

immunological response within the same patient, have not been performed.

This project aims to characterise and identify the predominant bacteria and

viruses in the nasopharynx, adenoids and the middle ear of children RAOM and

COME, for which surgical treatment, ventilation tube insertion (grommets), was

required. Furthermore, the presence of microbial biofilms in the middle ear, which

perform a pivotal role in antibiotic treatment failure and recurrent and/or chronic

or unresolving OM in these children will be explored. This project will also

28

characterise the host immune responses to these otopathogens. Host

immunological parameters, including innate parameters (cytokines) and adaptive

parameters (humoral and cellular immune responses) will be identified in each

participant’s middle ear effusion, saliva and serum. Finally, the project aims to

characterise the systemic immunology linked gene expression in peripheral

mononuclear cells in these otitis-prone children.

This is an observational study to characterise the otopathogens and immune

responses in urban children with RAOM/COME. The novel aspects of this project

relate to sampling from a comprehensive range of sites (middle ear, nasal

vestibule, nasopharynx, adenoids, blood, and saliva) within individual children to

permit closer examination of interactions between otopathogen presence and the

host’s immune responses. In addition, within the middle ear both biopsy and

effusion/wash samples will be taken.

Specific Project Aims

To identify microbes (bacteria and viruses) and microbial load in the

nasopharynx, adenoids and middle ears (left and right) in urban children

undergoing VTI for OM and in nasopharynx and adenoids in children

undergoing surgery for adenoid hypertrophy/obstructive sleep apnea

(AH/OSA) as control.

To characterise and compare specific serotypes or genetic types of

predominant bacteria, Spn and Hi, in the middle ear, nasopharynx and

adenoids in the two groups of children.

To examine bacterial biofilms in the middle ear samples of children with

OM.

To examine innate immune responses using selected cytokines, including

pro-inflammatory cytokines (IL-1β, IL-5, IL-6, IL-8, IL12, IL17, IFN-γ and

TNFα) and anti-inflammatory cytokines (IL10 and IL13), in the middle ear

and serum.

To examine adaptive immune responses including antigenic specific IgA,

IgG antibodies against four pneumococcal proteins (PspA1, PspA2, Ply,

and CbpA) and four Hi proteins (P26, P4, P6 and PD) in the middle ear,

saliva and serum.

29

To examine the potential role of Treg (CD4+CD25+FoxP3+) cells from

peripheral blood mononuclear cells in regulation of the host immune

response.

To investigate systemic immune gene expression.

To explore associations between bacterial species and viruses present in

the nasopharynx, the adenoids, and middle ear and host immune

responses in children with AH/OSA and children experiencing recurrent

acute otitis media or chronic otitis media with effusion (RAOM/COME).

30

Chapter 2:

General methods

31

2.1 Study populations and recruitment

All children recruited to the studies in this project were aged between 6 months

and 8 years of age, who were undergoing surgery for VTI or AH/OSA. Fully

informed consent was obtained from the child’s parent or guardian prior to

surgery and any study procedures being performed. Human Research Ethics

Committee approvals for the project, were obtained from Griffith University and

the treating hospitals prior to commencement of recruitment and are detailed for

each cohort below.

Exclusion criteria for all subjects recruited were any immunological abnormality,

intrinsic or pharmacological; anatomical or physiological defect of the ear;

syndromes associated with OM (e.g. cleft palate, Down syndrome);

sensorineurial hearing loss; and chronic mastoiditis, cholesteatoma or other OM

complications except for conductive hearing loss.

2.1.1 Pilot project: OM1 cohort

The recruitment of the pilot cohort (OM1) was approved by the Children’s Health

Services District Ethics Committee (EC00175, reference: 2008/063) and the

Griffith University Human Research Ethics Committee (EC00162, reference:

MSC/05/08/HREC).

Recruitment took place at the Royal Children’s Hospital. Informed consent was

taken from the parent/guardian prior to obtaining demographic information and

sample collection during surgery. Samples collected during surgery included a

<1.5mm3 biopsy of middle ear mucosa (MEM), middle ear effusion (MEE)

samples, a nasopharyngeal swab (NPS) and adenoid tissues. Saliva was

collected prior to anaesthesia (Figure 2-1).

OM group: children aged 1–7 years who had been referred to an ENT (Ear,

Nose, and Throat Surgeon) clinic for ventilation tube insertion (VTI) and

adenoidectomy for the treatment of RAOM or COME were recruited.

Control group: There were no control children in the first cohort.

32

2.1.2 Primary study: OM2 cohort and control cohort

Human research ethics approval was obtained from the Children’s Health

Services District Ethics Committee (EC00175, reference: HREC/14/QRCH/33),

Greenslopes Hospital Human Research Ethics Committee (EC00161, reference:

14/18) and the Griffith University Human Research Ethics Committee (EC00162,

reference: MSC/19/13/HREC) prior to commencement of recruitment.

Parents/guardians of eligible children were informed of the study following their

referral to an ENT surgeon and provided with a study information sheet, consent

form and a demographic questionnaire. The samples collected are detailed for

each group below.

OM2 cohort: children aged 6 months - 8 years who were undergoing VTI and

adenoidectomy for the treatment of RAOM or COME were recruited. Samples

were collected whilst the child was under general anaesthesia and included a

<1.5mm3 MEM biopsy, MEE, a NPS, a nasal swab (NS), adenoid tissues, tonsil

tissues (if available) and 10ml blood. Saliva was collected prior to anaesthesia

(Figure 2-2).

Control cohort: children aged 6 months - 8 years who have been referred to an

ENT for adenoidectomy (+/- tonsillectomy) for treatment of adenoids hypertrophy

(AH) and/or obstructive sleep apnoea (OSA) or snoring and with no significant

history of COME or RAOM. Samples were collected while the child was under

general anaesthesia and included a NPS, a NS, adenoid and tonsil tissues (if

available) and 10ml of blood. Saliva was collected prior to anaesthesia (Figure

2-2).

33

Figure 2-1: Samples collected from the OM1 cohort for microbiological analyses.

Children with RAOM/COME

(1 – 7 years of age)

NPS

MEE (left and right ears)

MEM (left and right ears)

Adenoids

Bacteria

Viruses

Bacterial biofilm

Microbiological assays:

Culture

PCR

FISH/CLSM

34

Figure 2-2: Samples collected from the OM2 and control cohort patients for

microbiological and immunological analyses. Samples from the middle ear were

only collected from OM2 patients, not control group patients and these samples

are highlighted in bold.


OR AH/OSA

(6 months – 8 years of age)

NPS

NS



Adenoids

Saliva

Blood

Bacteria

Viruses

NPS

NS



Adenoids


Saliva

Blood (serum)

Cytokines

Antibodies (IgA, IgG)

Microbiological assays Immunological assays

Blood

Treg cells

Gene expression

35

2.2 Sample collection

2.2.1 Saliva

Saliva was collected from children prior to anaesthesia induction, using cellulose

eye spears (Defries, Victoria, Australia) which were held in the patient’s mouth,

primarily under the patient’s tongue for 2 minutes. The specimen was kept at 4-

8oC in a plastic biohazard bag and transported to the processing laboratory within

4 hours of collection. Saliva was centrifuged at 4000rpm for 15 minutes (4oC),

aliquoted and stored at -80oC until analysis.

2.2.2 Samples collected under anaesthesia

All blood, NPS, NS, MEE, MEM, adenoid and tonsil tissues were collected from

children whilst they were under general anaesthesia. After collection, all samples

were stored on ice in a plastic biohazard bag and processed at the laboratory

within 4 hours of collection.

2.2.2.1 Blood

Blood samples were collected prior to commencement of middle ear or

adenoid/tonsil surgery. Whole blood (10ml) was collected and 2ml was

immediately transferred to a serum clot tube (InterPath, Victoria, Australia) with

the remainder, approximately 8ml being transferred into a Lithium Heparin tube

(InterPath, Victoria, Australia) for isolation of peripheral blood mononuclear cells

(PBMC).

2.2.2.2 Nasopharyngeal swab

NPS was collected by insertion of a sterile flexible cotton swab (Medline, Illinois,

USA) through the nare into the nasopharyngeal space (swab depth was equal to

distance from nostrils to outer opening of the auditory canal of ear ~ 3 – 4 cm for

~3 seconds). The swab was placed into a sterile tube containing 1ml of Skim-

Milk-Tryptone-Glucose-Glycerol-Broth (STGGB) (Hare et al., 2015, Kaijalainen

and Palmu, 2015, Kaijalainen et al., 2004). Following transport to the laboratory

36

the sample was vigorously vortexed for 30 sec, aliquoted and stored at -80oC until

analysis.

2.2.2.3 Nasal swab

A NS was collected by insertion of a sterile flexible cotton swab (Medline, Illinois,

USA) into the nostril into the nasal vestibule (<3cm depth) before being rotated

several times against nasal wall. The swab was placed in a sterile tube containing

1ml of STGGB. Upon return to the laboratory, the sample was vortexed, aliquoted

and stored at -80oC until analysis.

Following the collection of blood, the NS and the NPS samples, the outer ear

canals were gently cleaned with saline to minimise traces of topical antibiotic

drops. MEE, MEM and adenoid tissue were collected as surgery progressed.

2.2.2.4 Middle ear effusion

An anterior-inferior myringotomy incision was made utilising the operating

microscope and the MEE sample was collected aseptically using a sterile

aspirator tip and line to a Kendall Argyle 40ml specimen trap (Tyco Health Care

Group LP Mansfield, MA, USA) for each ear. For each ear, the tip and tubing

system was rinsed with sterile saline to recover all of the MEE. Upon return to the

laboratory, sample volume was measured, aliquoted and stored at 80oC until

analysis.

2.2.2.5 Middle ear mucosa

A small (<1.5mm3) sample of the middle ear mucosa (MEM) was removed from

the middle ear (cochlear) promontory, proximal to the Eustachian tube using

micro cup forceps via the myringotomy incision. Each specimen was placed in a

sterile tube containing 1ml of sterile Hanks buffered saline solution (HBSS)

(Gibco, NY, USA).

In the laboratory, the MEM was fixed overnight in 4% paraformaldehyde in

phosphate buffered saline (PBS) (pH 7.2), washed three times with 1X PBS and

stored in 50% ethanol/PBS at -20°C prior to analysis.

37

2.2.2.6 Adenoid tissue

Adenoid tissues were removed using a curette and placed in a sterile tube

containing 2ml of HBSS. Each specimen was kept on ice and transported to the

laboratory within 4 hours of collection, where it was dissected in half. One half

was used for isolation of mononuclear cells and the other half fixed and stored

for fluorescent in situ hybridisation (FISH). The HBSS buffer containing the

specimen was aliquoted and stored at -80oC for RT-PCR and culture.

Tissue for FISH was fixed overnight in 4% paraformaldehyde in PBS (pH 7.2),

washed three times with PBS and stored in 50% ethanol/PBS at -20°C prior to

hybridisation.

2.3 Bacterial culture

2.3.1 Clinical samples

MEE, NS, NPS samples and HBSS buffers from the adenoid samples were

cultured using standard pathology laboratory techniques. Ten µl loops of the

sample were streaked onto 3 plates; 1 horse blood agar plate for haemolytic

bacteria, 1 chocolate agar plate for fastidious gram-negative bacteria and 1

colistin nalidixic acid (CNA) agar plate for gram-positive bacteria. The plates were

incubated at 37oC with 5% CO2 and inspected for signs of growth at 24 and 48

hours. Bacteria were semi-quantitatively scored as in Figure 2-3.

To confirm bacterial colony identification using standard pathology practice,

colonies isolated in the OM1, OM2 and control cohort, had bacterial identification

for Hi and Mcat confirmed using Matrix Assisted Laser Desorption/Ionization Time

of Flight Mass Spectrometry (MALDI-TOF MS) (VITEK MS, BIOMERIEUX,

France). Suspected pneumococcal colonies were tested with Optochin tests

using a 6-mm-diameter disk containing 5µg of optichin. If the optochin result was

14mm diameter susceptible, it was reported as Spn and if the result was <14mm

diameter, the colony was confirmed by MALDI-TOF MS. All confirmed

otopathogens were tested for antibiotic susceptibility and β-lactamase

production.

38

Figure 2-3: Plating sample for semi-quantitative of bacteria detection.

2.3.2 Antibiotic susceptibility assay

Antibiotic susceptibility assays were performed and interpreted following

European Committee on Antimicrobial Susceptibility Testing (EUCAST)

guidelines, which are currently used by Pathology Laboratories in Queensland,

Australia (Table 2-1).

Hi and Mcat isolates were tested for amoxicillin/clavulanic acid (AMC), amoxicillin

(AMP), ciprofloxacin (CIP), ceftriaxone (CRO), cotrimoxazol or

trimethoprim/sulfamethoxazole (SXT) and tetracycline (TET) susceptibility.

Spn isolates were tested for SXT, TET, oxacillin (OX), vancomycin (VA) and

erythromycin (E) susceptibility.

Production of β-lactamase by Hi and Mcat strains was tested using Nitrocefin test

(Oxoid, Thermo Scientific, Australia), as per manufacturer instructions. A

Whatman #1 filter paper was placed in a sterile petri disk and impregnated with

2 drops of Nitrocefin reagent. Then a loop of isolated colony was applied to the

impregnated paper and the result was read within 15 minutes. Confirmation of β-

+1. growth in the first streak area. +2. growth in the second streak area. +3. growth in the third streak area. +4. growth in the last streak area.

39

lactamase production was positive when the paper turned deep pink or red and

negative when the paper remained yellow (no colour change).

Table 2-1: Zone diameter breakpoints to determine specific antibiotic

susceptibility for the 3 main bacterial otopathogens.

Disk content

Zone diameter breakpoint

(mm)

Susceptible Resistant

H. influenzae

CRO 30 µg ≥ 30 < 30

TET 30 µg ≥ 25 < 22

CIP 5 µg ≥ 26 < 26

AMP 2 µg ≥ 16 < 16

SXT 25 µg a ≥ 23 < 20

AMC 3 µg b ≥ 15 < 15

M. catarrhalis

CRO 30 µg ≥ 24 < 21

TET 30 µg ≥ 28 < 25

CIP 5 µg ≥ 26 < 26

AMP 2 µg ≥ 16 < 16

SXT 25 µg* ≥ 18 < 15

AMC 3 µg** ≥ 19 < 19

S. pneumoniae

SXT 25 µg* ≥ 18 < 15

OX 1 µg ≥ 20 <20

VA 5 µg ≥ 16 < 16

TET 30 µg ≥ 25 ≤ 22

E 15 µg ≥ 22 < 19

a 1.25µg trimethoprim and 23.75µg sulfamethoxazole. Break points are expressed as the trimethoprim concentration. b 2 µg amoxicillin and 1µg clavulanic acid.

2.3.3 Standard bacterial control samples

Bacterial standards were obtained from American Type Culture Collection

(ATCC, USA). Alloiococcus otitis (ATCC-51267), Haemophilus haemolyticus

(ATCC-33390), Streptococcus pneumoniae (ATCC-49619), Moraxella catarrhalis

(ATCC-25238), Streptococcus pyogenes (ATCC-19615), Staphylococcus aureus

(ATCC-29213), Pseudomonas aeruginosa (ATCC-27853), Haemophilus

influenzae type a (ATCC-9006) and non-typeable Haemophilus influenzae

40

(ATCC-49247) were cultured following manufacturer’s instruction and stocks

were stored in media containing 20% glycerol (Sigma, Australia) at -80oC (Figure

2-5). These strains were used as pure cultures for positive/negative controls for

PCR, BacLight staining and FISH.

2.4 Viral PCR

A panel of 10 viruses, including influenza A virus (IAV), influenza B virus (IBV),

parainfluenza virus 1 (PIV1), parainfluenza virus 2 (PIV2), parainfluenza virus 3

(PIV3), adenovirus (ADV), human metapneumovirus (hMPV), respiratory

syncytial virus (RSV), human rhinovirus (HRV) and WU polyomavirus (WU) was

analysed in adenoid, NPS and MEE samples of the OM1 cohort.

A panel of 17 viruses, including 10 viruses in the above panel and 7 additional

viruses (enterovirus – EV, human bocavirus – HboV, KI polyomavirus – KI, 229E

coronavirus - 229E, HKU1 coronavirus – HKU1, NL63 coronavirus – NL63 and

OC43 coronavirus – OC43) was analysed in adenoid, NS, NPS and MEE

samples of the OM2 and control cohorts.

Viral PCR assays were performed at Queensland Paediatric Infectious Diseases

(QPID) laboratory, Children Health Service, Brisbane, Queensland, Australia.

2.4.1 Nucleic acid extraction for viral detection

For each sample, total nucleic acid was extracted from 200µl of sample using

QIAxtractor instrument and DX viral nucleic acid reagents (Qiagen, Australia)

according to manufacturer instructions. Samples were spiked with 104 copies

DNA of Equine Herpes virus 1 (EHV1) prior to nucleic acid extraction to measure

efficiency of the process as previously described (Bialasiewicz et al., 2009,

Rockett et al., 2013). For each run (96 extracts/run) including 92 samples and 4

negative controls, extracts were tested using a duplex real-time PCR assay for

EHV1 and human endogenous retrovirus 3 (ERV3), which was used for human

sample qualification (Yuan et al., 2001). The samples were considered to have

failed the EHV1 (either failed extraction or possessed PCR inhibitors) if the EHV1

real-time PCR cycle threshold (Ct) results for the individual samples were higher

than two standard deviations (SD) from the mean value of Ct results from all

41

samples running at the same time (Bialasiewicz et al., 2009). For instance, if

mean value of Ct results from all samples with SD was 30 2, a sample with Ct

result more than 35 was classified as a failure. Samples failing this procedure

were re-extracted. Only those samples which passed EHV1 detection and were

positive for ERV3 were used for target viral analyses.

2.4.2 Viral detection

Samples which passed nucleic acid extraction quality control testing were

analysed for the presence of target viruses using previously optimised and

described PCR assays. Viral screening assays included: IAV and IBV (Whiley

and Sloots, 2005); PIV1/2/3 (Lambert et al., 2007); ADV (Lambert et al., 2012);

hMPV (Matsuzaki et al., 2009); RSV (van Elden et al., 2004); HRV (Lu et al.,

2008); WU and KI (Dang et al., 2011); 229E, HKU1, NL63 and OC43 (van Elden

et al., 2004, Dare et al., 2007); EV (McErlean et al., 2008); and HboV (Tozer et

al., 2009). Amplification and detection were performed using Quantitect Probe

PCR Mix (Qiagen, Australia) and Rotorgene instruments (Qiagen, Australia).

2.5 Bacterial PCR

2.5.1 Optimisation of DNA extraction from gram-positive bacteria

Lysis of bacterial cell-wall is a crucial step in DNA extraction. Enzymatic methods

using enzymes, such as Lysozyme (Lys), Mutanolysin (Muta) and

Achromopeptidase (Achro), were used to obtain efficient DNA extraction from

bacteria (Ezaki and Suzuki, 1982, Kalia et al., 1999, Yuan et al., 2012),

particularly gram-positive bacteria, which have thick layers of peptidoglycans

(Silhavy et al., 2010).

In this study, we compared DNA concentrations from pure cultures, which were

pre-treated with different enzymes before DNA extractions using the QIAamp

DNA mini kit (Qiagen, Australia). DNA concentration was measured by Quantus

Fluoror dsDNA system (Promega, Australia) and Quantus Fluorometer

(Promega, Australia).

42

All pure cultures with OD600 ~ 0.5, which were measured using Spectronic 200

(Thermo Scientific, Australia), were used for DNA extraction. Two-hundred µl of

bacterial suspension was centrifuged at 8000rpm for 5 minutes. The pellet was

collected and resuspended in a 40µl mixture of enzyme containing 20mg/ml Lys

(Sigma, Australia) only, or with either 100KU/ml Achro (Sigma, Australia) or

10KU/ml Muta (Sigma, Australia), or resuspended in 40µl Tris-EDTA (TE) buffer

only (Sigma, Australia). The bacterial suspensions with or without enzyme(s)

were incubated at 37oC for 30 minutes before DNA extraction with QIAamp DNA

mini kit (Qiagen, Australia).

Results of this optimisation showed that pre-treatment of Spn and S. pyogenes

with mixture of Lys and Muta provided significantly better extraction as evidenced

by higher concentration of bacterial DNA, compared to other mixtures, while pre-

treatment of S. aureus with a mixture of Lys and Achro showed a significantly

higher concentration of bacterial DNA. In contrast, no significant differences

between pre-treatment methods in DNA extraction of gram-negative bacteria

including Hi and Mcat were observed (Figure 2-4).

As the current study focussed on the 3 main bacterial otopathogens (Spn, Mcat

and Hi), an enzymatic mixture of Lys and Muta were chosen for pre-treatment of

clinical samples for DNA extraction.

43

S. pneumoniae

TE o

nly

Lys on

ly

Lys +

Ach

ro

Lys +

Mut

a0.0

0.2

0.4

0.6

0.8

1.0 ****

**D

NA

co

nc

en

tra

tio

n n

g/u

l

H. influenzae

TE o

nly

Lys on

ly

Lys +

Ach

ro

Lys +

Mut

a0.0

0.5

1.0

1.5

2.0

DN

A c

on

ce

ntr

ati

on

ng

/ul

S. pyogenes

TE o

nly

Lys on

ly

Lys +

Ach

ro

Lys +

Mut

a0.0

0.5

1.0

1.5

2.0

2.5

3.0 ****

**

M. catarrhalis

TE o

nly

Lys on

ly

Lys +

Ach

ro

Lys +

Mut

a0.0

0.5

1.0

1.5

2.0

S. aureus

TE o

nly

Lys on

ly

Lys +

Ach

ro

Lys +

Mut

a0.0

0.2

0.4

0.6

0.8

1.0 **

DN

A c

on

ce

ntr

ati

on

ng

/ul

Figure 2-4: The impact of pre-treatment of bacteria with different enzymes or TE

on the concentration of extracted DNA. Differences in the concentrations of

extracted DNA from pure cultures of S. pneumoniae, S. pyogenes, S. aureus, H.

influenzae and M. catarrhalis. The values are mean SEM (3 replicates) (*

p<0.05; ** p<0.01). TE: Tris-EDTA; Lys: Lysozyme; Achro: Achromopeptidase;

Muta: Mutanolysin.

44

2.5.2 DNA extraction for bacterial detection

DNA was extracted from 300µl of all samples. Samples were centrifuged at

8000rpm for 5 minutes. Supernatants were aspirated and pellets collected and

resuspended in Lys and Muta as described before incubation at 37oC for 30

minutes. DNA was then extracted using QIAamp DNA mini kit (Qiagen, Australia)

according to manufacturer’s instructions.

2.5.3 Bacterial detection

Real-time PCR was used to detect the three main bacterial otopathogens Spn,

Hi and Mcat. Targets for the bacteria were as follows: Spn (autolysin gene –

LytA), Hi (Haemophilus protein D gene – hpd and L-fucose permease gene -

FucP), Mcat (outer membrane protein gene – copB). In addition, using hpd#3

target in this study to differentiate between H. influenzae and H. haemolyticus as

described previously (Binks et al., 2012). FucP target was used when samples

were culture positive for H. influenzae but PCR negative for the hpd#3 target to

identify those H. influenzae strains missing the hpd gene (Smith-Vaughan et al.,

2014, Price et al., 2015).

Real-time PCR was carried out using Platinum® Quantitative PCR SuperMix (Life

Technologies, Australia) in 25µl reactions consisting of: 1x SuperMix, 300nM

primers (forward (F) and reverse (R) primers) and 200nM probe (P).

Amplifications were performed on iQ5 thermocycler (Biorad, Australia) with

cycling program: 50oC for 2 minutes, 95oC for 2minutes and 40 cycles of 95oC for

15 seconds and 55-60oC for 30 seconds. All primers and probes used in the

current study were obtained from Life Technologies and Biosearch Technologies

(Table 2-2).

45

Table 2-2: Primers and probes for bacterial detection using real-time PCR.

Target Name 5’-3’ nucleotide sequence Ref.

Hi

(hpd#3)

F-primer GGTTAAATATGCCGATGGTGTTG (Binks et al., 2012)

R-primer TGCATCTTTACGCACGGTGTA

Probe [HEX]-TTGTGTACACTCCGT “T” GGTAAAAGAACTTGCAC-[SpC6]

Hi

(fucP)

F-primer GCCGCTTCTGAGGCTGG (Price et al., 2015)

R-primer AACGACATTACCAATCCGATGG

Probe [6-FAM]-TCCATTACTGTTTGAAATAC-[MGBNFQ]

Spn

(lytA)

F-primer TCTTACGCAATCTAGCAGATGAAGC (Marsh et al., 2012)

R-primer GTTGTTTGGTTGGTTATTCGTGC

Probe [6-FAM]-TTTGCCGAAAACGCTTGATACAGGG-[TAMRA]

Mcat

(copB)

F-primer GTGAGTGCCGCTTTACAACC (Marsh et al., 2012)

R-primer TGTATCGCCTGCCAAGACAA

Probe [6-FAM]-TGCTTTTGCAGCTGTTAGCCAGCCTAA-TAMRA]

A. otitidis

(16S rRNA)

F-primer CTACGCATTTCACCGCTACAC (Marsh et al., 2012)

R-primer GGGGAAGAACACGGATAGGA

S. aureus

(nuc)

F-primer CCCTGAAGCAAGTGCATTTAC (Smith-Vaughan et al., 2008)

R-primer CAAGCCTTGACGAACTAA AGC

Probe [6-FAM]- TACCATTTTTCCATCAGCATAAATATACGCTAAGC-[TAMRA]

2.5.4 Bacterial load using quantitative PCR

Quantitative PCR to determine bacterial loads from NPS, NS and adenoid

samples were performed as previously described (Smith-Vaughan et al., 2006)

with modification. Purified DNA of S. pneumoniae (ATCC49619), H. influenzae

(ATCC49247) and M. catarrhalis (ATCC49247), in the range 50fg to 500pg, were

used to standardise the values against a standard curve (Figure 2-5). DNA loads

were converted to bacterial cells per ml of sample assuming that 1pg of

pneumococcal DNA was equivalent to 447.4 bacterial cells (a genome size of

2Mb) whilst 1pg of Hi DNA and Mcat DNA was equivalent to 497.1 bacterial cells

(a genome size of 1.8Mb) (Smith-Vaughan et al., 2006, Hoskins et al., 2001, De

Chiara et al., 2014, Nguyen et al., 1999).

46

Figure 2-5: Representative standard curve of quantitative PCR for bacterial load

analysis.

2.6 S. pneumoniae serotyping

Pneumococcal serotyping was performed using a Pneumotest kit (Statens Serum

Institut - SSI, Denmark), which is based on the gold standard Quellung reaction

(Habib et al., 2014, Jauneikaite et al., 2015), and following manufacturer’s

instruction. One-µl of overnight culture (approximately 8 hours) in broth was

dispensed on a biological glass slide (Livingstone, Australia). An equal amount

of antisera was added and mixed with the bacteria. The mixture was covered

immediately with coverslip to avoid dry out. Bacteria-antisera reaction was

examined under a phase contrast microscope with 100x magnification and oil

immersion lens. The reaction is stable for half an hour when it remains moist.

A positive Quellung reaction occurs when type-specific antibody binds to the

capsule of the pneumococcus, leading to a change in its refractive index. When

viewed under a microscope, the bacteria appear ‘swollen’ and more visible (Hare

et al., 2009), compared to a negative reaction when no change occurs (Figure 2-

6).

The Pneumotest kit was designed to differentiate the 23 serogroups/types found

in the pneumococcal polysaccharide vaccine (PPV23 or Pneumovax®) from non-

PPV23 groups/types. The vaccine groups/types must be positive for one of the

y = -1.518ln(x) + 39.985R² = 0.9996

0

5

10

15

20

25

30

35

40

1 10 100 1000 10000 100000 1000000

Th

resh

old

cycle

Log of DNA concentration (fg/µl)

47

pooled series A-H and one of the pooled series P-T. Non-vaccine groups/types

react only with one pool in the series A-H (Table 2-3).

Table 2-3: Chessboard of Pneumotest kit for pneumococcal serotype

identification based on Quellung reaction.

Pool P Q R S T Non-vaccine groups/types

A 1 18 4 5 2

B 19 6 3 8

C 7 20 24, 31, 40

D 9 11 16

E 12 10 33 21, 39

F 17 22 27, 32, 41

H 14 23 15 13, 28

Each pool contains antisera for several pneumococcal serogroups/types. In order to identify specific groups/types, Quellung reaction must be positive for 2 pools of antisera. For example, pneumococcal serogroup 11 (bold-red) isolate must be positive for pools D and T. If an isolate is only positive for pool D and negative for both pools R and T, it will be classified as non-PPV23 serogroups and may belong to serogroup 16.

48

Figure 2-6: Representative images of positive and negative pneumococcal

antigen and antibody reactions of Quellung test (Skovsted, 2015). a) negative

reaction without swollen capsules; b) positive reaction with swollen capsules.

2.7 H. influenzae typing

H. influenzae was typed by PCR using a modified published protocol

(primers/probes listed in Table 2-4) (Wroblewski et al., 2013). H. influenzae,

which was detected using culture/MALDI-TOF and/or PCR, was identified

whether it was encapsulated or non-encapsulated by PCR with primers/probe for

bexA gene. If an isolate was negative for bexA, it was determined to be a non-

encapsulated strain or NTHi strain (Figure 2-7). If an isolate was positive for

bexA, it was considered to be an encapsulated strain and was subject to

identification with specific primers for specific types (Table 2-4).

Real time PCR was carried out using Platinum® Quantitative PCR SuperMix (Life

Technologies, Australia) in 25µl reactions consisting of: 1x SuperMix, 450nM

primers (forward (F) and reverse (R) primers) and 250nM probe (P).

Amplifications were performed on iQ5 thermocycler (Biorad, Australia) with

cycling program: 50oC for 2 minutes, 95oC for 2minutes and 40 cycles of 95oC for

15 seconds and 60oC for 40 seconds.

49

Table 2-4: Primers used for PCR typing H. influenzae.

Type Gene target

Primer Primer sequence (5′–3′) Amplicon size (bp)

All type

bexA BexA-F TCTTGGCCGTTAGCTTTTAGTGG

148 BexA-R CATAAAGATAATCGCCCAATTCAGA

BexA-P FAM-ACTCGTTTTACAAAAGAGTTT-TAMRA

a acsA AcsA-F GCAACCATCTTACAACTTAGCGAAT

80 AcsA-R CTGCGGTGTCCTGTGTTTAGG

AcsA-P FAM-CTATCAAGTGAAGCATGTCGCCATTCGTC-TAMRA

b bcsB BcsB-F CAAGATACCTTTGGTCGTCTGACTAA

151 BcsB-R TAGGCTCGAAGAATGAGAAGTTTTG

BcsB-P FAM-TGATATGGGTACATCTGTTCGCCATAACTTCATCTTAG-TAMRA

c ccsB CcsB-F TCTGCAAGAAATGTTGGAATTGA

130 CcsB-R GTGAAACATTCTCTGATTCTGATATTTGT

CcsB-P FAM-ACGTTCAAACCGAATGGGTTACTTTCATTGA-TAMRA

d dcsC DcsC-F TTGATGACCGATACAACCTGTTTAA

119 DcsC-R CACGAAATTATTTCTCCGTTATGTTGA

DcsC-P FAM-GTTTTACAACCATTCTTA-TAMRA

e ecsC EcsC-F GAGTTTAAGGATTTTATTGAGCTTTTCG

78 EcsC-R AGCAATAGTTTGAAAGAACCCTCTGT

EcsC-P FAM-TCAAGCCGCTCCACT-TAMRA

f fcsA FcsA-F CTTACGCCTGAAATTTGCTATTACTTT

105 FcsA-R AGAATGTGGTCTATTTCCATTCTCTTTT

FcsA-P FAM-ATATTCGCTCCGCAAATGCTAAGGCAC-TAMRA

50

Figure 2-7: Scheme for typing H. influenzae strains by PCR.

2.8 LIVE/DEAD staining of middle ear effusion

To identify live and dead bacteria and to determine morphological presence in

MEEs, LIVE/DEAD staining of MEEs were performed using a BacLight kit

(Invitrogen, USA) in accordance with manufacturer instruction and previously

published protocols (Thornton et al., 2013).

2.8.1 Testing reagent with standard bacteria

One ml of standard bacterial culture with OD600 ~ 0.5 was used to test the

BacLight staining kit. Bacterial cultures were centrifuged at 8000rpm for 5 minutes

and the supernatant was removed. The bacterial pellet was washed twice with

500µl of 0.85% saline and centrifuged at 8000rpm for 5 minutes each. The final

pellet was resuspended in 350µl of 0.85% saline for staining.

Fifty µl of bacterial suspension was air dried on SuperFrost Plus slides (Menzel-

Glaser, Australia) at room temperature in biological hood. BacLight working

solution was made by combining 0.75µl of SYTO9 and 0.75µl of propidium iodide

(PI) in sterile saline following manufacturer’s instruction. Samples were incubated

with 50µl of BacLight working solution in darkness for 30 minutes at room

temperature. After incubation, samples were washed 3 times for 5 minutes each

with 1X PBS. Samples were mounted using a low fade mounting media, covered

H. influenzae from positive samples

Non-typeable/unencapsulated

H. influenzae

Typeable/encapsulated

H. influenzae

Type a or b or c or

d or e or f

-bexA

+bexA

+acsA or +bcsB or +ccsB or

+dcsC or +ecsC or +fcsA

51

by coverslip and sealed with nail polish. The stained specimens were stored in

darkness at -20oC until imaging.

Stained slides were assessed using Confocal Laser Scanning Microscopy

(CLSM) with standard FITC filter sets (Excitation/Emission: 480/500) for SYTO®9

and Texas Red® (Excitation/Emission: 490/635) for PI (Figure 2-8).

52

Figure 2-8: Pure culture of ATCC H. influenzae was stained with LIVE/DEAD

BacLight reagents and analysed under CSLM. a) SYTO9 stain (green) labels all

bacteria (live and dead); b) PI stain (red) only labels damaged cell-wall bacteria

(dead bacteria); c) Merged image of a and b showing live and dead bacteria.

Scale bars are 10µm.

10µm 10µm

10µm

53

2.8.2 Sample staining

MEE samples from children with RAOM/COME were examined using

LIVE/DEAD staining to examine the viability and morphological organisation of

bacteria. Fifty µl of the MEE samples in saline were air dried on SuperFrost Plus

slides at room temperature for 30 minutes. BacLight working solution was made

by combining 0.75µl of SYTO9 and 0.75µl of PI in sterile saline following

manufacturer’s instruction. Samples were incubated with 50 µl of BacLight

working solution in darkness for 30 minutes at room temperature. Following

incubation, samples were washed 3 times for 5 minutes each with 1X PBS.

Samples were mounted using a low fade mounting media, covered by coverslip

and sealed using nail polish. The stained specimens were stored in darkness at

-20oC until imaging.

Stained samples were analysed using CLSM, an optical imaging technique used

to obtain high resolution images on relatively thick, fully hydrated samples. CLSM

is fluorescent microscopy based and uses a pinhole to remove out of focus light

or flare from specimens, which is apparent in other fluorescent microscopy.

Samples were scored positive when imaging results demonstrated bacterial

morphology (size: from 0.5µm to 2µm; shape: either cocci or cocci-bacilli), biofilm

structure and fluorescence with appropriate signal. In contrast, samples were

scored negative when no bacterial morphologies were observed.

2.9 Fluorescent in situ Hybridisation

To identify the specific bacteria present within the middle ear mucosal biopsy

(MEM), FISH was used. This is a method by which fluorescently labelled

oligonucleotide probes target and bind to rRNA allowing specific identification

bacterial species (Speel, 1999). Species specific probes were used in

accordance with previously published protocol (Thornton et al., 2011). The

sequences for these probes are found below in Table 2-5.

2.9.1 Specificity of probes

Oligonucleotide probes, labelled with Alexa Fluor (AF) (Molecular probes, USA),

were designed to detect specific bacteria based on 16S rRNA sequences (Hall-

54

Stoodley et al., 2006, Thornton et al., 2011) (Table 2-5). Probe specificity was

tested using panels of standard bacteria. One-hundred µl of standard bacteria

cultures with OD600 ~ 0.5 were mixed together. The bacteria mixtures were fixed

with 4% formaldehyde in PBS overnight at 4oC and washed with 1X PBS 3 times.

The fixed bacteria mixture was stored in 50% ethanol/PBS at -20oC until

analysed.

Fifty µl of fixed bacterial mixture was air dried on SuperFrost Plus slides at room

temperature for 30min. Bacteria were treated with Lysozyme (Sigma, Australia)

in a hybridisation chamber for 2 hours at 37oC for permeabilisation. Before

staining, bacteria were dehydrated with series of ethanol concentrations (50%,

80% and 100%) for 3 minutes each. Bacteria were then hybridised with specific

probes for target bacteria within a hybridisation chamber with corresponding

formamide concentrations in hybridisation buffer at 46oC for 2 hours. For each

sample 2ml of hybridisation buffer was made up as described in Table 2.6.

Following hybridisation, bacteria were washed with pre-warmed washing buffer

(48oC) appropriate to the formamide concentration used (Table 2-7), for 15

minutes in a water bath (48oC). Finally, nucleic acid of hybridized bacteria were

stained with 1.2µg/ml Hoechst 33342 (Thermo Fisher, Australia) in darkness at

4oC overnight prior to bacteria were imaged. Samples were stored at -20oC until

imaging with CLSM.

Specificity of bacterial and non-bacterial probes were showed in series of images

below (Figures 2-9, 2-10, 2-11, 2-12).

55

Table 2-5: Probes used to identify bacterial species using FISH.

Probes Target Sequences (5’-3’) Formamide concentration

EUB338-AF546

Bacteria

(16S[338-355])

GCTGCCTCCCGTAGGAGT 20% (0 – 40%)

NONEUB338-AF488

Non-bacteria

ACT CCT ACG GGA GGC AGC

20%

Spn-AF488 S. pneumoniae

(16S[195-212])

GTGATGCAAGTGCACCTT 20%

Hi-AF633 H. influenzae

(16S[185-202])

CCGCACTTTCATCTCCG 20%

Mcat-AF488 M. catarrhalis

(16S[88-105])

CCGCCACUAAGUAUCAGA 10%

Table 2.6: In situ hybridisation buffer.

5M NaCl 360µl

1M Tris-HCl, pH 8.0 40µl

Formamide X*µl

Baxters water Added to 2ml

10% SDS 2µl

* The volume depends on probes used to obtain the appropriate formamide concentration.

Table 2-7: Washing buffer appropriate to formamide concentration used in

hybridisation buffer.

% formamide in hybridisation buffer

1M Tris-HCl (µl)

5M NaCl (µl)

0.5M EDTA (µl)

Baxters water

10% SDS (µl)

10 1000 4500 - Fill to 50ml 50

20 1000 2150 500 Fill to 50ml 50

56

Figure 2-9: Mixture of all standard bacteria were stained with Hoechst 33342 for

DNA, NONEUB338-AF488 probe (non-bacterial probe) as a negative control and

EUB338-AF546 probe, which is a universal bacterial probe, as a positive control.

a. Nucleic acid of bacteria were stained with Hoechst 33342 (blue) b. No bacteria

were detected in channel with NONEUB338-AF488 probe (green) c. Unidentified

bacteria were detected in channel with EUB338-AF546 probe (red) d. merged

image of a, b and c. Scale bars are 20µm.

57

Figure 2-10: Mixture of standard bacteria without Spn were stained with Hoechst


evaluate the specificity of Spn-AF488 probe. a) Nucleic acid of bacteria were

stained with Hoechst 33342 (blue); b) Spn was not detected in channel with Spn-

AF488 probe (green), demonstrating the probe cannot detect other tested

bacteria; c) unidentified bacteria were detected in channel with EUB338-AF546

probe (red); d) Hi was detected in channel with Hi-AF633 probe (grey); e) Merged

image of a, b, c and d. Scale bars are 20µm.

58

Figure 2-11: Mixture of standard bacteria without Hi were stained with Hoechst


evaluate the specificity of Hi-AF633 probe. a) Nucleic acid of bacteria were

stained with Hoechst 33342 (blue); b) Spn was detected in channel with Spn-

AF488 probe (green); c) Unidentified bacteria were detected in channel with

EUB338-AF546 probe (red); d) Hi was not detected in channel with Hi-AF633

probe (grey), demonstrating the probe cannot detect other tested bacteria; e)

Merged image of a, b, c and d. Scale bars are 20µm.

59

Figure 2-12: Mixture of all standard bacteria including Spn and Hi were stained

with Hoechst 33342, Spn-AF488 probe, EUB338-AF546 probe and Hi-AF633

probe to assess multiple probes used in one sample. a) Nucleic acid of bacteria

were stained with Hoechst 33342 (blue); b) Spn was detected in channel with

Spn-AF488 probe (green); c) Bacteria were detected in channel with EUB338-

AF546 probe (red); d) Hi was detected in channel with Hi-AF633 (grey), which

was different from Spn detected in (a), demonstrating two different probes can

detected 2 different targets in one sample; e) Merged image of a, b, c and d.

Scale bars are 20µm.

60

2.9.2 FISH of middle ear biopsy

To identify the specific bacteria within the MEM, the interaction between them,

and their position on the mucosal specimens, FISH was performed.

Analysis of MEM samples was performed in accordance with previously

published protocols (Hall-Stoodley et al., 2006, Thornton et al., 2011). Middle ear

biopsy samples were processed on non-charged, non-treated slides. Samples

were dehydrated in an ethanol dilution series for 3 min intervals at concentrations

of 50%, 80% and 100% ethanol. As the bacterial composition in the samples are

unknown and they likely contain gram positive organisms, samples were then

permeabilised with Lysozyme at 37oC for 2 hours in a hybridisation chamber to

maintain humidity. Following this, the biopsy samples were dehydrated using the

ethanol dilution series as previously described. Following dehydration, the

samples were then hybridised with specific probes for target bacteria within a

hybridisation chamber with corresponding formamide concentrations in

hybridisation buffer at 46oC for 2 hours. For each sample 2ml of hybridisation

buffer was made up as described in Table 2.6. Following hybridisation, biopsies

were washed with pre-warmed washing buffer (48oC) appropriate to the

formamide concentration used (Table 2-7), for 15 minutes in a water bath (48oC).

Finally, nucleic acid of hybridized biopsies were stained in 1ml PBS containing

1.2µg/ml Hoechst 33342 in darkness at 4oC overnight prior to the biopsies were

imaged.

Specimens were imaged by CLSM. Image collection was performed sequentially

to ensure only emission from the stimulated fluorophore was being detected,

thereby improving the imaging specificity. Samples were scored positive when

results demonstrated correct bacterial morphology (size: from 0.5µm to 2µm;

shape: either cocci or cocci-bacilli), biofilm structure and fluorescence with

appropriate channel.

61

2.10 Serum collection

Whole blood (2ml) was collected into a clot tube (InterPath, Victoria, Australia)

allowed to clot for a minimum of 30 minutes and transferred to the laboratory

within 4 hours. The sample was centrifuged at 3220g for 10 minutes. Following

centrifugation the serum was aspirated from the separated sample (Figure 2-13),

aliquoted and stored at -80oC until analysis.

Figure 2-13: Serum collection from whole blood using centrifugation.

2.11 IgA and IgG assays

Multiplex fluorescent bead assays, established and validated at the University of

Western Australia, Australia were used to quantitatively detect IgG and IgA

antibodies to pneumococcal proteins PspA1, PspA2, Ply, and CbpA and Hi

proteins P26, P4, P6, and PD in accordance with a previously published method

(Wiertsema et al., 2012) but modified to include the P26 which was recently

established in the laboratory. Samples were measured using a 3-plex assay

including microspheres conjugated with PspA1, P26 and PD, and a 5-plex assay

including microspheres conjugated with PspA2, CbpA, Ply, P4 and P6. The

BioPlex® 200 System (Biorad, Australia) and Bio-plex Manager 5.0 software

were used for analysis. The lower limitations for detection for the protein assays

are determined as following results (AU/ml): P4 IgA (0.44); P4 IgG (0.13); P6

IgA/IgG (0.13); P26 IgA (0.76); P26 IgG (0.13); PD IgA (0.14); PD IgG (0.13);

62

CbpA IgA (0.13); CbpA IgG (0.13); PspA1 IgA/IgG (0.13); PspA2 IgA/IgG (0.13)

and Ply IgA/IgG (0.13).

Total protein in MEE and saliva samples was determined by the Pierce BCA

(Bicinchoninic acid) Protein Assay kit (Rockford) according to manufacturer’s

instructions. Samples were measured at an initial dilution of 1:30 and modified if

they fell out of the range of the standard curve. Antibody titres in MEE and saliva

samples were calculated per milligram (mg) of total protein per millilitre (ml) of

samples to standardise results to accommodate sample collection effects.

Serum antibody titres were reported in arbitrary units (AU) per ml (AU/ml). MEE

and saliva antibody titres were reported in AU per mg of total protein per ml of

sample (AU/mg).

2.12 Cytokine assays

A panel of 10 cytokines (IL-1β, IL-5, IL-6, IL-8, IL-10, IL-12p70, IL-13, IL-17, IFN-

γ and IFN-α), which was established and validated by The University of Western

Australia, Australia, was performed to measure cytokine levels in serum and MEE

samples. The absorbance of each sample was measured using BioPlex® 200

System (Biorad, Australia). Sensitivities for the cytokine assays are determined

for each cytokine as following results: 2.44pg/ml for IL-5, IL-1B, IL-6, IL12-p70,

IFNg, TNFa and IL-10; 4.88pg/ml for IL-8, IL-13 and IL-17.

MEE cytokine data were normalised with total protein each sample to standardise

results to accommodate sample collection effects and reported in picogram (pg)

per mg of total protein per ml of sample (pg/mg).

2.13 Peripheral blood mononuclear cell isolation

Eight ml of whole blood was drawn into a Lithium heparin tube (InterPath, Victoria,

Australia) inverted several times and transferred to the laboratory within 4 hours

of collection. The blood was mixed with an equal volume of 1X PBS and gently

layered over 10ml of ficoll (GE healthcare, USA) in a 50ml tube. The layered

sample was the centrifuged at 400g for 30 minutes (with no brake).

63

Figure 2-14: PBMC isolation from whole blood using Ficoll separation.

Following centrifugation (Figure 2-14) the PBMC layer (buffy coat), was

harvested and collected into a separate tube. PBMC were then washed two times

with 10ml of 1X PBS by centrifuge at 250g for 10 minutes. After washing steps,

the cells were resuspended in 5ml of 1x PBS and counted by NucleoCounter NC-

200 (Chemometec AS, Australia). Following counting, the cells were centrifuged

and resuspended in FBS containing 10% DMSO to a final concentration of 3-5 x

106 cells/ml. PBMC in cryovials were placed in a Mr Frosty (Nalgene, Australia)

and frozen down overnight at -80oC. Vials were then transferred to liquid nitrogen

(LN2) for long term storage as previously published method (Mallone et al., 2011).

2.14 Regulatory T cell assay

Regulatory T cells (Treg cells) from PBMC were analysed by flow cytometry

(Fortessa, BD, Australia) to detect surface markers (CD4 and CD25) and the

intracellular marker (FoxP3), which are specific markers for Treg cells (Buckner,

2010).

2.14.1 Thawing cryopreserved PBMC

PBMC were previously stored in FBS containing 10% DMSO in liquid nitrogen.

Cells were thawed and washed as described previously (Sattui et al., 2012) with

modification. The cells in 1ml cryovial were transferred immediately from LN2 to

37oC incubator. Thawed cells were transferred to 4ml of pre-warmed FBS and

64

centrifuged at 400g for 5 minutes. The cell pellet was collected and resuspended

in 2ml of FBS before centrifugation at 400g for 5 minutes. The supernatant was

aspirated and pellet was resuspended in 5ml of PBS containing 10% FBS. The

cell suspension was counted and viability checked using the NucleoCounter

(Chemometec AS, Australia). Only samples with over 90% viability were used for

cell marker staining.

To assess differences between fresh and frozen PBMC with and without resting

in FBS overnight at 37oC and 5% CO2 after thawing on cell viability and recovery

of viable cells, we used PBMC from adult volunteers and analysed the cells 1)

immediately after the cell isolation 2) after being frozen in LN2 for 4 weeks without

resting overnight post thawing and 3) after being frozen in LN2 for 4 weeks with

resting post thawing.

Viability of PBMC was significantly higher in fresh PBMC when compared to

thawed PBMC without resting (p=0.04; Figure 2-15a). No differences in PBMC

viability was observed between those that were thawed and rested and either

fresh cells or thawed PBMC without resting (Figure 2-15a). However, when

numbers of viable cells that were recovered were assessed, PBMC that had been

frozen, thawed and then rested were reduced by 30-50% compared to fresh

PBMC and by 20-30% when compared to thawed PBMC without resting (Figure

2-15b). Taken together, thawed PBMC without resting were chosen for T cell

analyses due to viability after thawing and without resting was still higher than

90% and number of viable cell recovery was higher than from thawed cells with

overnight resting.

65

a. cell viability

b. viable cell recovery

Figure 2-15: Comparison cell viability and viable cell recovery between fresh

PBMC, thawed PBMC with and without resting overnight. Student’s t tests were

performed for comparison between groups.

Thawed PBMC without resting

Thawed PBMC with

resting

Fresh PBMC

Thawed PBMC without resting

Thawed PBMC with

resting

Fresh PBMC

Number of viable cells

66

2.14.2 Cell staining for flow cytometry

Cells with more than 90% of viability were adjusted to have concentration

between 1 and 5x106 cells/ml by stain buffer (BD, Australia). A 100µl aliquot of

cells suspension was transferred to a 5ml polystyrene tube (BD falcon, Australia).

Each sample had one tube for unstained cells or fluorescence minus one; one

tube for CD4, CD25 staining and isotype of FoxP3; 2 tubes for CD4, CD25 and

FoxP3 staining. Cells were firstly stained with antibodies for surface markers

(FITC-CD4 and BV421-CD25) as per manufacturer recommended

concentrations (BD, Australia) at 2-8oC in the dark for 25 minutes. After

incubation, the cells were washed with 2ml of stain buffer then centrifuged at 400g

for 5 minutes. Cells were then fixed and permeabilised by resuspension of the

pellet with 1ml of 1X Fix/Perm solution (BD, Australia) followed by a further 40–

50 minutes incubation at 2-8oC in darkness. Cells were then washed again, twice,

using 2ml of 1X Perm/Wash (BD, Australia) then centrifuged at 400g for 5

minutes. The cells were stained with FoxP3 antibody or Ig isotype using

manufacturer recommended concentrations (BD, Australia) and incubated for

40–50 minutes in darkness at 2-8oC. After intracellular staining, the cells were

washed twice using 2ml of 1X Perm/Wash (BD, Australia), centrifuged at 400g

for 5 minutes and then the stained cells pellet was resuspended in 200µl of stain

buffer and stored at 2-8oC in the dark until analysis by flow cytometry. A

diagrammatic representation of the staining procedure is shown in Figure 2-16.

67

Figure 2-16: Overview of Treg cell staining procedure adapted from BD protocol

(BD Pharmingen, 2015). PBMC were firstly stained with antibodies surface

markers (CD4 and CD25). The cells then were washed, fixed and permeabilised

before they were stained with antibody of intracellular marker (FoxP3). After

intracellular marker staining, the cells were washed and analysed immediately or

stored in dark at 4oC until analysed.

68

2.14.3 Flow cytometry analysis

2.14.3.1 Setting up and controls

Flow cytometry was performed using BD LSRFortessa cell analyser (BD,

Australia), which can detect up to 18 colours simultaneously, supporting up to 4

lasers and analysed using BD FACSDiva software (BD, Australia) and FlowJo

software (FlowJo, USA).

Quality assurance for consistent and reproducible data was performed at every

use of the cell analyser, using BD cytometer setup and tracking (CST) system

and CST beads (BD, Australia). In addition, the BD™ CompBeads Set Anti-

Mouse Ig, κ (BD, Australia) were used to set up and optimise fluorescence

compensation or remove fluorescence spillover for multicolour flow cytometric

analysis of Treg cells. These polystyrene microparticles were stained with

fluorescent antibodies using the previously described procedure for staining

PBMC. Nonspecific staining of the antibodies was determined using isotype

controls, conjugated to particular fluorochromes. Gating controls were performed

using fluorescence minus one (FMO) controls, which include all of the antibody

conjugates present in the test except one. Results of isotype controls showed

very small percentage of non-specific staining or signals compared to specific

antibodies for CD4, CD25 and FoxP3 (Figure 2-17). Data from isotype controls

were used to normalise final data of each samples.

69

FMO and isotype controls for CD4 gating using FITC-CD4 antibody

FMO and isotype controls CD25 gating using BV421-CD25 antibody

FMO and isotype controls FoxP3 gating using AF647-FoxP3 antibody

Figure 2-17: Setting controls and gating for each subset of T lymphocyte using

FMO and isotype. FMO controls for gating cell population; Isotype controls for

identifying nonspecific signals.

70

2.14.3.2 Gating strategy

The strategy for gating Treg cells from PBMC was illustrated in Figure 2-18. A

lymphocyte gate was established around lymphocytes in a forward and side

scatter plot (a). From this plot, cells positive for CD4 were determined in (b). Cells

positive for both CD4 and CD25 were identified as CD4+CD25high cells, which

had fluorescence intensity more than 2x102, and CD4+CD25 low cells, which had

fluorescence intensity less than 2x102 in (c). Finally, Treg cells, which were

positive for CD25 (high or low) and FoxP3, were determined in (d) and (e). This

strategy was performed using FlowJo software.

71

Figure 2-18: Gating strategy for Treg cells from PBMC.

a

b

c

d

e

72

2.15 Gene expression assay

The method for gene expression is based on Nanostring technologies, which was

previously described (Geiss et al., 2008, Cesano, 2015).

2.15.1 Cell lysate preparation

Cell lysates containing the RNA from PBMC were used to analyse gene

expression. Frozen PBMC were thawed at 37oC for 10 minutes. Samples were

centrifuged at 400g for 5 minutes and the cell pellet was collected. The pellet was

then washed with 5ml of 1x PBS and centrifuged at 400g for 5 minutes. Following

washing the pellet was resuspended in 1ml of 1x PBS and the cells was counted

and checked viability using automated cell analyser N-200 (Chemometec AS,

Australia). Cells with a viability of more than 90% were used for lysis. A total of

200,000 cells were lysed by adding 20µl of RLT buffer (Qiagen, Australia) and

vortexing for 1 minute. The cell lysate was spun down to recover all material

before aliquoting and storage at -80oC until analysis.

2.15.2 RNA hybridisation

The nCounter XT gene expression assay (Nanostring, USA) was used and the

hybridisation procedure was performed following manufacturer instructions. A

mastermix for a 12-tube strip was created by adding 70µl of hybridisation buffer

and 49µl of RNase-free water to the tube containing the Reporter CodeSet and

then mixed by inverting repeatedly and rapidly spinning down. The hybridisation

mix was created by adding 11.5µl of the mastermix and 1.5µl of the cell lysate to

each tube. Finally, 2µl of Capture ProbeSet was added to the hybridisation mix

and mixed by inverting several times before placed in a pre-heated 65oC thermal

cycler. The hybridisation mix was incubated at 65oC for 16 hours.

2.15.3 Reaction analysis

Reaction analysis was performed by nCounter Platform including Prep Station

and Digital Analyser (Nanostring, USA). After hybridisation, excess probes were

removed automatically by the machine and probe/target complexes aligned and

immobilised in the nCounter Cartridge by the Prep Station. The sample Cartridge

73

was then placed in the Digital Analyser for data collection. Colour codes on the

surface of the cartridge were counted and tabulated for each target molecule.

2.15.4 Data analysis

The nCounter XT assay is a multiplex assay for 579 known genes involved in

human immune responses. The assay also contains probes for 15 internal

reference genes with different levels of expression to evaluate quality of each

sample. Positive and negative controls are included in the assay to assess

performance of the assay and the platform. The raw Nanostring gene expression

data were normalised using negative controls, positive controls and

housekeeping genes via nSolver version 2.6 software (Nanostring, USA).

Normalised data was then analysed for functional enrichment and detection of

significant pathways using Ingenuity Pathway Analysis software (IPA, Qiagen,

USA) as previously described (Krämer et al., 2014). Prior to be analysed by IPA,

genes with more than 10 counts after normalisation were considered as

expression and cut-off values ( 1.5 or ≤ -1.5) of fold change were used for

analysis. The principal method embedded in IPA for biological functions/pathway

enrichment is singular enrichment analysis. The programme searched through

the list of selected genes and determined genes that involved in the respective

Knowledge Base, including gene ontology (GO) term, canonical signal

transduction or metabolic pathways. Then, Fisher's Exact algorithm was used to

calculate the probability of which each functional gene set was enrichment. Only

the biological functions/pathways (Bonferroni's corrected p<0.05) were

considered significant enrichment. The activation status of the

functions/pathways were predicted using IPA Upstream Regulator Analysis Tool

by calculating a regulation Z-score and an overlap p-value, which were based on

the number of known target genes of interest pathway/function, expression

changes of these target genes and their agreement with literature findings. It was

considered significantly activated (or inhibited) with an overlap p<0.05 and an IPA

activation Z-score ≥ 2.0 (or ≤ −2.0). The detailed descriptions of IPA analysis are

available under “Upstream Regulator Analysis”, “Biological Functions Analysis”,

and “Ingenuity Canonical Pathways Analysis” on the IPA website

(http://www.ingenuity.com).

74

2.16 Statistical analysis

Statistical analyses for each individual analyses were described in each result

chapter. Overall, p<0.05 was considered statistically significant. The IBM SPSS

Statistics 23 for Windows software package (IBM, New York, USA) was used for

all statistical analyses and data were plotted using GraphPad Prism 5.0 (Graph

Software Inc., California, USA).

Power analysis to calculate the potential group sizes needed to determine

significance between groups was calculated as described below. Unfortunately,

optimum group sizes were not achieved within the final recruitment phase of this

project cohort, which significantly limited some aspects of the study, however the

use of paired samples from within individuals provided additional, within individual

controls.

The sample size was calculated using microbiological information/data from

previously published study (Wiertsema et al., 2011b, Wiertsema et al., 2011a),

which was conducted in Western Australia and had similar study population.

Details sample size calculation was performed as below:

Sample size with 80% (β = 0.2) power and 5% significance (α = 0.05) for 2 groups

of children with (group 1) and without (group 2) RAOM/COME was calculated

using following formula:

α = 0.05 => Z2α = 1.96

β = 0.2 => Z2β = 1.04

p1: hypothetical proportion of specific microbe positive in nasopharynx from group 1

p2: hypothetical proportion of specific microbe positive in nasopharynx from group 2

q1 = 1 – p1

q2 = 1 – p2

p = (p1 + p2)/2

q = (q1 + q2)/2

75

Table 2-8: Sample size for each group based on results from previous studies in

Western Australia (Wiertsema et al., 2011b, Wiertsema et al., 2011a).

Specific microbe positive in the nasopharynx

p1 p2 Sample size for each group needed (n)

S. pneumoniae 41% 26% 158

H. influenza 56% 19% 31

M. catarrhalis 43% 40% 3986

Rhinovirus 58% 42% 155

RSV 28% 9% 71

Adenovirus 25% 6% 63

Bocavirus 52% 20% 40

Parainfluenza virus 29% 9% 66

76

Chapter 3:

Bacterial otopathogens in the upper

respiratory tract and middle ear of

children with and without RAOM/COME in

South East Queensland

77

3.1 Introduction

S. pneumoniae (Spn), H. influenzae (Hi) and M. catarrhalis (Mcat) are the three

main bacterial otopathogens causal for otitis media (OM) (Massa et al., 2009).

Colonisation of the nasopharynx and adenoids within the upper respiratory tract

by these bacteria is associated with development of recurrent acute OM (RAOM)

and chronic OM with effusion (COME) (Harabuchi et al., 1994, Faden et al., 1997,

Saylam et al., 2010). Despite high rates of detection of these bacteria in the

nasopharynx and adenoids using bacterial culture, middle ear effusion (MEE)

samples are often culture-negative. Culture negative samples may result from

low bacterial loads, bacterial establishment and survival in biofilms or through

intracellular infection (Hall-Stoodley et al., 2006, Coates et al., 2008, Thornton et

al., 2011).

Molecular techniques, such as polymerase chain reaction (PCR), provide a

sensitive method for detection of non-cultureable bacteria in MEE (Post et al.,

1995, Hendolin et al., 1997). Overall, the average detection frequencies of

bacterial otopathogens in MEE samples analysed by PCR are 3 times higher than

detections analysed using culture as reported in a systematic review (Ngo et al.,

2016). This is consistent with the 1-4 fold (Marsh et al., 2012, Wiertsema et al.,

2011b) improvement reported from previous reports examining samples from

Indigenous and non-Indigenous Australian children. In this study, PCR and

culture were used to identify predominant bacteria within the middle ear and other

commonly sampled upper respiratory tract (URT) locations, in children with and

without RAOM/COME.

Large bacterial loads colonising the nose and nasopharynx are associated with

persistence of MEE and progression to suppurative OM in Aboriginal children in

remote areas of Australia (Smith-Vaughan et al., 2006, Binks et al., 2011). OM

occurrence and severity in children from these regions is significantly greater than

that observed for urban children. This study examined bacterial loads of

predominant bacteria within the URT of urban children, both with and without

RAOM/COME to determine the impact of bacterial loads in OM pathogenesis in

children from urban areas of Eastern Australia.

78

Although urban and remote environments clearly differ, there are environmental

factors, such as attending day-care, having siblings, being exposed to cigarette

smoke and not being breastfed, which may contribute to high rates of bacterial

transmission, which in turn may cause OM (Rovers et al., 2004, Lehmann et al.,

2008). OM prevalence in children from remote Australian communities is

influenced by these environmental factors (Jacoby et al., 2008, Jacoby et al.,

2011). This study aimed to compare a range of environmental factors between

children with and without RAOM/COME recruited from urban South East

Queensland to determine their potential association with OM aetiology in the

region.

Environmental conditions as described above, also include vaccination programs

that may impact OM pathogenesis. For example, the heptavalent pneumococcal

conjugate vaccine (PCV7), contains 7 pneumococcal serotypes (4, 6B, 9V, 14,

18C, 19F and 23F) and when introduced into National Immunisation Programme

(NIP), reduced the rate of invasive pneumococcal diseases (Tan, 2012).

Importantly, PCV7 also showed efficacy against vaccine serotype OM, but did not

result in an overall reduction of OM incidence. Replacement disease with non-

PCV7-serotypes and other bacterial otopathogens are hypothesised to explain

the minimal impact of PCV7 on OM incidence (Eskola et al., 2001, Casey et al.,

2010, Dupont et al., 2010). In Australia, PCV7 was included in the NIP in 2005

and reduction in OM caused by PCV7 serotypes was subsequently observed.

Concurrently, the detection of non-PCV7 serotypes of Spn and of Hi were

increased in children with and without RAOM (Wiertsema et al., 2011b).

Replacement of PCV7 in the NIP with a 13-valent pneumococcal conjugate

vaccine (PCV13) occurred in 2011. Compared to PCV7, PCV13 included an

additional 6 pneumococcal serotypes (1, 3, 5, 6A, 7F and 19A) (Johnson et al.,

2012, Collins et al., 2013). Currently, there are no published reports on the

efficacy and impact of PCV13 on OM incidence in young children within Australia.

Recruitment of children to this study occurred over the period of PCV13 inclusion

in the NIP and thus permitted examination of both PCV7 and PCV13 vaccinated

children, with or without RAOM/COME.

79

Overall, this study aimed to examine URT carriage rates of the 3 main bacteria

causal for OM in urban children to determine the bacterial aetiology of

RAOM/COME in Eastern Australia. Nasopharyngeal, nasal, adenoid and MEE

samples were collected from children with RAOM/COME and were compared to

results from nasopharyngeal, nasal and adenoid samples collected from control

group children without RAOM/COME. The URT location, identification of co-

colonisation of predominant bacteria and antibiotic sensitivity of identified

bacteria were examined in control and RAOM/COME children to explore bacterial

transference between locations of the URT. Specific bacterial identification of

predominant bacteria causal for OM was performed using bacterial culture, PCR

and imaging techniques. Imaging techniques were used to demonstrate bacterial

location within the middle ear, in either biofilms or as intracellular bacteria within

the epithelial cells as these persistence mechanisms may contribute to the

recurrence and chronicity of OM in these children. Improving our understanding

of the patterns of co-colonisation of the different bacteria predominantly causal

for OM in the URT of urban children may guide future development of OM

prevention strategies including vaccine development and use of probiotics.

3.2 Methods

Patient recruitment, sample collection and processing, bacterial culture and

antibiotic susceptibility testing, PCR methodology, bacterial typing, bacterial

LIVE/DEAD staining and fluorescent in situ hybridisation (FISH) were performed

as described in Chapter 2, sections 1, 2, 3, 5, 6, 7, 8, and 9, respectively. To

differentiate H. influenzae and H. haemolyticus, real-time PCR with

primers/probes for hpd#3 and/or fucP was used as described in Chapter 2 as well

as in previous studies (Binks et al., 2012, Price et al., 2015).

Data were analysed using the statistical analysis package SPSS for Windows,

version 23. Host and environmental risk factors were compared between control

group children, who have no significant history of OM, and children with

RAOM/COME. Categorical variables, including gender and vaccination status

were assessed using Pearson Chi-square analyses (p-value asymptotic

significant 2-sided). Age and number of episodes of OM were compared using

independent samples t tests. Upper respiratory tract carriage of bacteria, patterns

80

of co-colonising or infecting species, and antibiotic susceptibility patterns were

compared between OM and control cohorts using Pearson Chi-square analyses.

McNemar’s tests were used for comparison of detection frequencies between

PCR and culture results. For all statistical analyses, a p-value <0.05 was

considered significant.

81

3.3 Results

3.3.1 Study population

Forty-three children with a history of at least 4 episodes of AOM within the last 12

months (RAOM) or persistence of MEE for longer than 3 months (COME),

undergoing ventilation tube insertion (VTI) as treatment for OM were recruited.

Children were otherwise healthy as assessed by anaesthetists on day of surgery.

An initial pilot study of 23 children, who were fully vaccinated with PCV7, was

recruited between December 2008 to May 2010 to form OM cohort 1 (OM1). After

expansion of the URT sample collection sites, a further 20 children were recruited

between April and November 2015, and who were vaccinated with either PCV7

(n=9) or PCV13 (n=11), formed the OM cohort 2 (OM2). All 43 children had the

contents of their middle ears aspirated (MEE) and a small biopsy removed (MEM)

at the time of VTI. In total 85 MEE samples were collected, since one child in

OM1 had a spontaneous tympanic membrane perforation identified at the time of

surgery which was not sampled since only MEE from intact middle ears was

collected.

A control group of 17 children, without a history of RAOM or COME, who were

otherwise healthy and undergoing surgery for adenoidal hypertrophy (AH) or

obstructive sleep apnoea (OSA), were recruited. All children recruited to this

study were well at the time of surgery and underwent general anaesthesia during

which samples were collected.

Characteristic demographics of these patient cohorts are presented in Table 3-

1. All children were recruited in the urban areas of South East Queensland and

no Indigenous Australian children were enrolled in this study.

82


cohort enrolled in this study.

OM1 OM2 Control p-value a

p-value b

Number 23 20 17

Mean age in year (range) 3.3 (1-7) 4.1 (2–8) 4.4 (3–8) 0.16 0.60

Male (%) 12 (52%) 11 (55%) 12 (71%) 0.71 0.51

Mean episodes of AOM within last 12 months (range)

>4c 7.1 (4-11) 0.9 (0-2) na <0.001

Attending day-care na 18/19 (95%)

13/16 (81%)

na 0.21

Antibiotic use in the last month

na 11/19 (58%)

7/16 (44%)

na 0.40

Having siblings na 17/19 (89%)

15/16 (94%)

na 0.65

A history of breastfeeding na 16/19 (84%)

14/16 (88%)

na 0.78

A history of pacifier use na 11/19 (58%)

7/16 (44%)

na 0.40

Allergy to any substance d na 4e/19 (21%)

4f/16 (25%)

na 0.78

Asthma na 2/19 (11%) 4/16 (25%)

na 0.26

Exposed to cigarette smoke

na 6/19 (32%) 4/16 (25%)

na 0.67

Fully vaccinated with PCV7

23 (100%)

9 (45%) 9 (53%) 0.00 0.63


0 (0%) 11 (55%) 8 (47%) 0.00 0.63

a OM1 vs OM2 b OM2 vs Control c The specific frequency of AOM episodes not available for OM1 cohort d Cow’s milk protein, egg, wheat, shellfish, house dust mite, pollens/trees, moulds, grasses, bee/insect venom, peanut, other nuts, cat/dog hair, others. e Allergic to penicillin (n=2), strawberries and kiwi fruits (n=1) and cow’s milk protein (n=1). f Allergic to penicillin (n=1), wheat (n=1), gluten (n=1) and cow’s milk protein (n=1).

na: not available

Student’s t tests were used for continuous variables (age and number episodes of OM)

Pearson Chi-Square analyses were used categorical variables (Male, attending day-care, antibiotics use, siblings, breastfeeding, pacifier use, allergy, asthma cigarette smoke expose and vaccination)

83

3.3.2 Bacterial carriage

Upper respiratory tract carriage of the three predominant bacterial otopathogens

(Spn, Hi and Mcat) was examined using bacterial culture and/or PCR in children

from the OM1, OM2 and control (AH/OSA) groups. The three predominant

bacteria identified within each region of the URT are described for each location,

individually, below.

3.3.2.1 Bacteria identified within the adenoids of children with

RAOM/COME compared to control

All bacteria detected using culture were also detected using PCR, in all adenoid

samples. At least one otopathogen (Spn or Hi or Mcat) was detected in all

adenoids collected from all children, with and without RAOM/COME (Table 3-2).

No significant difference in the detection frequency of specific bacteria was

observed between the RAOM/COME (OM) and control groups. H. influenzae was

the predominant bacterium detected within the adenoid tissues in the OM2 (90%)

and control cohort (94%) samples (p=0.69, Table 3-2). Spn and Mcat were also

frequently detected in adenoid tissues from children in the OM2 (65%, 75%) and

control cohorts (65%, 69%), respectively, however, the detection frequencies did

not differ between the control and RAOM/COME patient groups (p=0.52 for Spn

and p=0.81 for Mcat; Table 3-2).

Concurrent detection frequencies of two bacteria only, such as Spn and Hi, or Hi

and Mcat or Spn and Mcat, within the same adenoid sample, did not differ

significantly between the OM2 and control groups (Table 3.2).The frequency of

detection for the presence of all three bacteria (Spn, Hi and Mcat) within the

adenoid samples did not differ between the OM2 (40%) and control (50%) cohorts

(p=0.48; Table 3-2).

The bacteria detected as a single otopathogen within the adenoid samples was

Hi. Hi was detected as a single bacterium in OM2 (15%) and control (13%) groups

respectively. In comparison, Mcat identified as a single pathogen occurred within

only a single OM2 sample, whilst Spn was never detected in isolation from the

other predominant bacteria in adenoids of either group of children (Table 3-2).

84

Detection frequencies of the predominant bacterial otopathogens within the

adenoids did not differ in association with vaccination with either PCV7 or PCV13.

(Table 3-3).

85

Table 3-2: Bacterial otopathogens detection from the adenoids, nasopharynx (NPS), nasal (NS) and middle ear (MEE) of children

with RAOM/COME (OM1 and OM2) and children without RAOM/COME (control).

Adenoids NPS NS MEE

Bacteria OM2

(n = 20)

Control

(n = 16)

p-value

OM1

(n = 23)

OM2

(n = 20)

Control

(n = 17)

p-value a p-value b OM2

(n = 20)

Control

(n = 17)

p-value

OM1

(n = 45)

OM2

(n = 40)

p-value

Children positive for at least one otopathogen by PCR

20 (100%) 16 (100%) 1.00 18 (78%) # 19 (95%)

14 (82%)

0.11 0.22 15 (75%)

10 (59%) # 0.46 17 (38%) # 24 (60%) # 0.01

Children positive for at least one otopathogen by Culture

15 (75%) 12 (75%) 1.00 9 (41%)1 15 (75%)

12 (71%)

0.02 0.76 10 (50%)

3 (18%) 0.04 4 (9%)2 1 (3%) 0.21

Spn by PCR

Spn by culture

13 (65%)

10 (50%)

12 (75%)

8 (50%)

0.52

1.00

14 (61%) #

6 (27%)

6 (30%)

5 (25%)

7 (41%)

5 (29%)

0.04

0.94

0.17

0.76

9 (45%)

4 (20%)

6 (35%)

3 (18%)

0.55

0.86

9 (20%) #

3 (7%)

6 (15%)

0 (0.0%)

0.55

0.10

Hi by PCR

Hi by culture

18 (90%) #

7 (35%)

15 (94%) #

5 (31%)

0.69

0.81

16 (70%) #

6 (27%)

13 (65%)

8 (40%)

10 (59%)

6 (35%)

0.75

0.33

0.67

0.77

8 (40%) #

2 (10%)

5 (29%)

2 (12%)

0.50

0.86

13 (29%) #

1 (2%)

10 (25%) #

1 (3%)

0.86

0.56

Mcat by PCR

Mcat by culture

13 (65% ) #

5 (25%)

11 (69%) #

3 (19%)

0.81

0.65

11 (48%) #

4 (18%)

11 (55%)

9 (45%)

8 (47%)

8 (47%)

0.43

0.05

0.43

0.90

10 (50%)

6 (30%)

4 (24%)

2 (12%)

0.10

0.10

7 (16%)

0 (0%)

10 (25%)

0 (0%)

0.28

0.29

Only one bacterial otopathogen (PCR only)

Spn

Hi

Mcat

0 (0%)

3 (15%)

1 (5%)

0 (0%)

2 (13%)

0 (0%)

n/a

0.83

0.36

2 (9%)

3 (13%

0 (0%)

0 (0%)

7 (35%)

3 (15%)

2 (12%)

3 (18%)

2 (12%)

0.18

0.09

0.05

0.13

0.24

0.77

0 (0%)

2 (10%)

2 (10%)

3 (18%)

2 (12%)

2 (12%)

0.05

0.86

0.86

2 (4%)

5 (11%)

1 (2%)

6 (15%)

8 (20%)

8 (20%)

0.10

0.14

0.008

Two of three bacterial otopathogens (PCR only)

Spn + Hi

Spn + Mcat

Hi + Mcat

4 (20%)

1 (5%)

3 (15%)

3 (19%)

1 (6%)

2 (13%)

0.93

0.87

0.83

2 (9%)

0 (0%)

1 (4%)

0 (0%)

3 (15%)

3 (15%)

1 (6%)

0 (0%)

2 (12%)

0.18

0.05

0.23

0.27

0.10

0.77

2 (10%)

5 (25%)

1 (5%)

1 (6%)

0 (0%)

0 (0%)

0.65

0.03

0.35

3 (7%)

1 (2%)

2 (4%)

0 (0%)

0 (0%)

2 (5%)

0.10

0.34

0.90

All three bacterial otopathogens (PCR only)

8 (40%) 8 (50%) 0.49 10 (44%) 3 (15%) 4 (24%) 0.07 0.51 2 (10%) 2 (12%) 0.86 3 (7%) 0 (0%) 0.10

a OM1 vs OM2; b OM2 vs Control. # Significant difference between PCR and culture results using McNemar’s test (p<0.05); Pearson Chi-square was used to compare frequencies of detection in each of the specimens.1 OM1 NPS samples cultured (n=22); 2 OM1 MEE samples cultured (n=43).

86

Table 3-3: Bacterial otopathogens identified in the upper respiratory tract of

children fully vaccinated with either PCV7 or PCV13 within the OM and control

groups.

Fully PCV7 vaccinated

Fully PCV13 vaccinated

p-value

OM2 cohort – Adenoid samples

Spn 7/9 (78%) 6/11 (55%) 0.28 Hi 8/9 (89%) 10/11 (91%) 0.88

Mcat 5/9 (56%) 8/11 (73%) 0.42

Control cohort – Adenoid samples

Spn 5/8 (63%) 7/8 (88%) 0.25 Hi 7/8 (88%) 8/8 (100%) 0.30

Mcat 6/8 (75%) 5/8 (63%) 0.59

OM cohorts – NPS samples

Spn 16/32 (50%) 4/11 (36%) 0.43 Hi 22/32 (69%) 7/11 (64%) 0.76

Mcat 14/32 (44%) 8/11 (73%) 0.14

Control cohorts – NPS samples

Spn 2/8 (25%) 5/9 (56%) 0.20 Hi 4/8 (50%) 6/9 (67%) 0.49

Mcat 5/8 (63%) 3/9 (33%) 0.23

OM2 cohort – NS samples

Spn 5/9 (56%) 4/11 (36%) 0.39 Hi 5/9 (56%) 2/11 (27%) 0.20

Mcat 3/9 (33%) 7/11 (64%) 0.18

Control cohort – NS samples

Spn 2/8 (25%) 4/9 (44%) 0.40 Hi 2/8 (25%) 3/9 (33%) 0.71

Mcat 1/8 (13%) 3/9 (33%) 0.31

OM cohorts – MEE samples

Spn 13/63 (21%) 2/22 (9%) 0.22 Hi 17/63 (27%) 5/22 (23%) 0.70

Mcat 8/63 (13%) 9/22 (41%) 0.004

Values represent the number of samples positive for that bacterium compared to the total number of samples tested with the percentage of positive samples indicated in the bracket. Comparison of the percentage of positive samples of each bacterium within the different locations was performed using Pearson Chi-square analyses.

87

3.3.2.2 Bacterial detection in the nasopharynx of children with

RAOM/COME compared to controls

For all nasopharyngeal samples, all bacteria detected in the samples using

culture were also detected using PCR. Within the nasopharynx, there were high

detection rates (PCR) for identification of at least one of the three predominant

bacterial otopathogens in the OM1 (78%), OM2 (95%) and control (82%) groups,

respectively (Table 3-2). These detection rates did not differ significantly between

the recruited patient groups (OM1 vs OM2; p=0.11 and OM2 vs control; p=0.22;

Table 3-2). Bacterial otopathogen detection frequencies were higher using PCR

versus culture methods for detection, for all groups recruited. Furthermore, the

OM1 group showed significantly lower rates of bacterial detection using culture

than were observed in the OM2 and Control groups (p<0.05, McNemar’s test)

compared to the PCR analyses of the same specimens.

Detection frequencies of specific otopathogens in the nasopharynx did not differ

between children from the OM2 and control cohorts (Table 3-2). In the OM1

group, however, Spn detections (PCR) were significantly higher than that

observed for OM2, 61% versus 30% respectively (p=0.04), whilst Mcat detections

(culture) were lower (18% versus 45%) respectively.

Hi was the most common otopathogen in all cohorts, however there were no

significant differences in the frequency of detection between the patient groups

(PCR, p=0.67; culture, p=0.77; Table 3-2). The frequencies of Hi detected in the

NPS using PCR were 70%, 65% and 59% in the nasopharynx of children from

the OM1, OM2 and control cohorts, respectively.

Spn was the second most common bacterium in the nasopharynx of children from

the OM1 cohort and it was detected less frequently in the nasopharynx of children

from the OM2 cohort (61% vs 30%, p=0.04; Table 3-2). Spn detection in the

nasopharynx of children from the OM2 and control cohorts did not differ

significantly (30% vs 41%, p=0.17; Table 3-2). Although Mcat detection

frequency in the NPS did not differ between the three cohorts analysed using

PCR, culture based analyses showed increased detection frequency in the OM2

cohort compared to the OM1 (45% vs 18%, respectively; Table 3-2).

88

Co-detection of Spn and Hi occurred at low frequency in the nasopharynx of

children from the OM1, OM2 and control cohorts (9%, 0% and 6% respectively;

Table 3-2). Similarly, the frequency of co-detection of Hi and Mcat in the same

samples was also low 4%, 15% and 12% in the nasopharynx of children from the

OM1, OM2 and control cohorts, respectively (Table 3-2). Frequency of Spn and

Mcat detected together in the nasopharynx of children from the OM2 was higher

than that of children from the OM1 and control cohorts but the differences were

not statistically significant (p=0.054 for OM1 vs OM2; p=0.10 for OM2 vs control;

Table 3-2).

Co-detection for the three main bacteria within the nasopharynx was highest in

children from the OM1 cohort but not significantly different to the frequencies

observed in the other two (p=0.07 for OM1 vs OM2, p=0.51 for OM2 vs control;

Table 3-2).

Spn was not detected in isolation from other otopathogens in the nasopharynx of

children from the OM2 cohort but was identified as a single pathogen in 9% and

12% of the nasopharynx of children from the OM1 and control cohorts,

respectively (p=0.18 for OM1 vs OM2; p=0.12 for OM2 vs control; Table 3-2). Hi

was detected in isolation from other bacteria in 13%, 35% and 18% of the

nasopharynx from the OM1, OM2 and control cohorts, respectively (p=0.09 for

OM1 and OM2; p=0.24 for OM2 and control; Table 3-2). Mcat was not detected

alone in the nasopharynx of children from the OM1 cohort but in 15% and 12%

of children from the OM2 and control cohorts, respectively (p=0.77; Table 3-2).

The frequency of detection of otopathogenic bacteria within the nasopharynx did

not differ in association with vaccination status (PCV7 or PCV13). Furthermore,

specific frequency of detection of Hi, Spn and Mcat did not vary significantly in

the NPS samples from PCV7 and PCV13 immunised children. Children with

RAOM/COME and PCV13 vaccinated showed consistent, albeit not significant,

reduction in Spn and increase in Mcat detections for all sample locations including

the nasopharynx compared to PCV7-vaccinated children in the OM cohorts,

respectively (36% vs 50%, p=0.43 for Spn; 73% vs 44%, p=0.14 for Mcat; Table

3-3).

89

3.3.2.3 Bacterial detection from the nasal vestibule of children with

RAOM/COME compared to controls

Nasal samples (NS) collected from the vestibule of the nose, were only sampled

from the OM2 and control patients. All otopathogenic bacteria detected by culture

were also detected by PCR. The detection frequency of children who were

positive for at least one bacterial otopathogen (by culture) was significantly higher

in children with OM, compared to control children (p=0.04; Table 3-2) but this

difference was not observed using PCR (p=0.46; Table 3-2). All subsequent

description and comparison of results are described using the PCR data.

Detection frequencies for Spn, Hi and Mcat within the nasal vestibule were slightly

but not significantly higher in the OM2 group compared to control (Table 3.2).

When detected as single bacterial otopathogens, in isolation from other

pathogens, Hi and Mcat were detected at similar rates between the control (12%)

and OM2 group (10%) (p=0.86) (Table 3-2), whilst Spn was not detected as a

single bacterial pathogen in the nose in the OM2 group. Although Hi and Mcat

were the most frequently detected as a single pathogen (10%) in the OM2 group,

Spn was the most frequently identified single pathogen in the control group (18%)

(Table 3-2).

Multiple bacterial otopathogens were detected, using culture, in 50% and 18% of

NS samples from children in the OM and control groups respectively (p=0.04;

Table 3-2), however there was no significant difference using PCR (p=0.46;

Table 3-2). All 3 bacterial pathogens were co-detected in the NS of children from

the OM2 (10%) and Control (12%) groups (p=0.86; Table 3-2). Spn and Mcat

were the bacteria most frequently co-detected in the NS of children in the OM2

group (25%), compared to controls (0%) (p=0.03; Table 3-2).

Vaccination status, whether fully immunised using PCV-7 or PCV-13 did not

significantly alter the frequencies of detection of the predominant bacterial

otopathogens in the nose, in the OM2 and control cohorts. No significant

differences in detection of Hi, Spn and Mcat were observed between children with

or without RAOM/COME. Spn and Hi detection frequencies were slightly but not

90

significantly lower whilst Mcat detection was slightly but not significantly higher in

PCV13-vaccinated children in the OM2 cohort compared to PCV7-vaccinated

children in the OM2 cohort, respectively (Table 3-3).

3.3.2.4 Bacterial detection from the MEE of children with RAOM/COME

compared to controls

Overall, forty-five MEE samples from the OM1 cohort and 40 MEE samples from

the OM2 cohort were examined and all bacterial otopathogens detected by

culture were also detected by PCR. PCR results were used for subsequent

comparisons described in this section.

In the OM1 cohort, Hi was detected most often (29%), followed by Spn (20%) and

Mcat (16%) (Table 3-2), with Hi, Spn and Mcat detected as single pathogens in

11%, 4% and 2% of MEE samples respectively. Co-detection with at least one

other bacterial pathogen occurred at similar frequencies for Hi (18%), Spn (16%)

and Mcat (13%) (p=0.43; Table 3.2).

In the OM2 cohort, Hi and Mcat were detected at the same frequency (25%) in

comparison to Spn (15%) (p=0.46; Table 3-2), with Hi, Mcat and Spn detections

as the sole otopathogen shown in 20%, 20% and 15% of MEE samples,

respectively. Co-detection of Mcat and Hi occurred in 5% of samples, whilst Spn

was not detected concurrently with any other bacterial pathogen in the MEE

(Table 3-2).

When bacterial results from both OM cohorts, OM1 and OM2, were compared,

single otopathogen detections were more frequent in the OM2 MEE samples

(55% vs 16%, p=0.000; Table 3-2). Of those pathogens that were detected in

isolation, Mcat was identified more frequently in the OM2 cohort (20% vs 2%,

p=0.008; Table 3-2), whilst co-infection with at least 2 of the pathogens was

observed more frequently in the MEE from OM1 (20% vs 5%, p=0.04; Table 3-

2).

Vaccination status, whether fully immunised using PCV-7 or PCV-13 did not show

significant differences in Hi and Spn detections within the middle ear (p=0.70 and

p=0.22, respectively) (Table 3.3). In contrast, immunisation with PCV-13 was

91

associated with increased detection frequency for Mcat within the MEE compared

to children vaccinated with PCV-7 (41% vs 13%, p=0.004; Table 3-3).

To explore the potential microbiological relationship between an individual’s two

middle ear cavities, 42 bilateral MEE samples (left and right ears) were examined

(OM1=22 children, OM2=20 children). In total, 62% of children (26/42) had at

least one bacterial otopathogen detected from either one or both ears (Table 3-

4) and of these children, 6 children (23%) had the same otopathogen detected in

both ears, whilst 20 children (77%) had different bacteria detected in their left and

right ears.

92

Table 3-4: Individual pathogen detection in the middle ears of children with OM

that had at least one middle ear positive for bacterial otopathogens.

Cohort Patients Ear Hi Spn Mcat

OM1 Child No. 2 Left - - -

Right + - -

OM1 Child No. 4 Left - - +

Right + - +

OM1 Child No. 5 Left + - -

Right - - -

OM1 Child No.13 Left - - -

Right - + +

OM1 Child No.15 Left + + +

Right + + -

OM1 Child No.18 Left + + -

Right + - -

OM1 Child No.20 Left + + +

Right + + -

OM1 Child No.22 Left - + -

Right - + -

OM1 Child No.23 Left + - +

Right + + +


Right - + -

OM2 Child No. 5 Left - + -

Right - - -


Right + - -


Right - - -


Right - - +


Right - - +


Right - + -


Right - - +


Right - + -


Right + - -

OM2 Child No. 23 Left + - +

Right + - -


Right + - +


Right + - -


Right + - -


Right - - +


Right - - +


Right - - -

93

3.3.3 Relationship of bacterial detections in sampling sites of the

upper respiratory tract

Identification of the same bacteria in different URT locations within each individual

patient was examined using PCR-positive and PCR-negative results from NS,

NPS and adenoid samples from the control group children and from MEE, NS,

NPS and adenoid samples from the children in the OM cohorts. Chi-Square

analyses were performed to evaluate whether association between different URT

locations was significant (p<0.05).

Detection of each of the 3 main bacterial otopathogens within the nasal vestibule

was associated with concurrent detection in the nasopharynx of the same

children, regardless of OM history (p=0.003 for Spn, p=0.000 for Hi and Mcat).

Furthermore, detection of Hi and/or Mcat within the NS was always associated

with detection of the same bacteria within the NPS. In contrast, Spn detected in

the NS was not always concurrently detected in the NPS of the same child (33%)

(Figure 3-1).

Similarly, clear associations for the detection of the 3 main bacterial otopathogens

were observed between the nasopharynx and adenoids of children regardless of

OM history (p=0.02 for Spn and Hi, p=0.000 for Mcat). All Hi or Mcat detections

within the nasopharynx, were also detected in the adenoids of the same child. In

contrast, Spn detected in the nasopharynx was not always concurrently detected

in the adenoids of the same child (8%) (Figure 3-1).

The potential relationship/s between detection of bacterial otopathogens in the

middle ear and other URT locations within the same child was performed after

combination of the bacterial detections from both ears as a single combined

sample.

Detection of Spn observed within MEE samples was significantly associated with

detections within other URT locations (MEE vs NS, p=0.02; MEE vs NPS, p=0.02

(OM1) and p=0.00 (OM2); MEE vs adenoids, p=0.03). Spn detections in the

middle ear from children in the OM1 (100%) and OM2 (83%) cohorts, were also

detected in the nasopharynx of the same child (Figure 3-1). Similarly, middle ear

94

detection of Spn was always associated with detection of Spn in the adenoids

and 83% of Spn detected in the nasal vestibule of the same child (Figure 3-1).

Detection of Hi and Mcat in the middle ear was not always significantly associated

with detection in other URT locations within the same children (Hi: MEE vs NS,

p=0.25; MEE vs NPS, p=0.69 (OM1) and p=0.17 (OM2); MEE vs adenoids,

p=0.29; Mcat: MEE vs NS, p=0.16; MEE vs NPS, p=0.006 (OM1) and p=0.09

(OM2); MEE vs adenoids, p=0.17). Overall, detections of Hi (75-86%, 100%) and

Mcat (86-100%, 86%) within the middle ear and were also detected in the

nasopharynx or adenoids of the same child, respectively (Figure 3-1). In contrast,

only 57% of Hi and 71% of Mcat detected in the middle ear were also detected in

the nasal vestibule (Figure 3-1).

95

NPS

Adenoids

NS

MEE

Spn

Hi

Mcat

Spn

Hi

Mcat

Spn

Hi

Mcat Spn

Hi

Mcat

Hi (100%, 15) Mcat (100%, 15) Spn (67%, 15)

Hi (100%, 20) Mcat (100%, 19) Spn (92%, 12)

Spn (100%, 6)

Hi (100%, 7) Mcat (86%, 7)

Mcat (86-100%, 12) Spn (83-100%, 12) Hi (75-86%, 15)

Spn (83%, 6) Mcat (71%, 7) Hi (57%, 7)

96

Figure 3-1: Co-detection of bacterial otopathogens in different locations of the

upper respiratory tract from children with and without RAOM/COME.

This figure represents the bacteria identified using PCR in the different upper

respiratory tract (URT) locations for all children, with and without a history of

RAOM/COME and the associated locations where the same bacteria were

identified in the same child.

Each circle represents an URT location and for each of the three bacteria, the

percentage of total bacteria identified in the location and the number of children

are shown (x%, y); where more than one cohort of OM patients were included ex.

MEE samples, a range of percentages is shown (x%-z%, y). The size of the

circles illustrating each URT location are representative of the total number of

bacterial detections (ex. NPS circle is larger than NS; more bacterial detections

in NPS).

Association of bacterial detections in one location compared to another are

illustrated using dash arrows. In children where the bacteria detected in one area

were also detected in another area, this is demonstrated in the figure. For

example, bold red Spn (100%, 6) indicates that all Spn detected in middle ear

were also detected in adenoids of the same child and 6 MEE samples positive

for Spn were compared to Spn detection results in 6 matching adenoidal samples

from the same children. Blue dash line indicates data from the OM2 and control

cohorts; Green dash line indicates data from the OM2 cohort only; Orange dash

line indicates data from the OM1 and OM2 cohorts.

97

3.3.4 Co-occurrence of bacterial otopathogens in the upper

respiratory tract of children with and without RAOM/COME

Co-occurrence of three main bacteria in the nasopharynx and the adenoids of

children with and without RAOM/COME showed that for children with OM, only

Spn and Mcat were associated in the NPS (r=0.445, p=0.003; Table 3-5). NPS

samples from control group children showed no significant co-occurrence of any

of the 3 main otopathogens (Table 3-5) and no significant association was

observed between bacterial otopathogens identified in the adenoids of children

from the OM or control cohorts (Table 3-5).

98

Table 3-5: Co-occurrence of the three main otopathogens in the nasopharynx

and the adenoids of children from the OM and Control cohorts.

N With Spn, N (%) With Mcat, N (%)

The nasopharynx

OM1 and OM2 cohorts

Hi Pos Hi Neg p-value

29 16

13 (45%) 5 (31%)

0.32

17 (59%) 6 (38%) 0.331

Spn Pos Spn Neg p-value

22 23

16 (73%) 7 (30%) 0.001

Control cohort


10 7

4 (40%) 3 (43%)

0.38

6 (60%) 2 (25%)

0.20


7 10

4 (57%) 4 (40%)

0.49

The adenoids

OM2 cohort

Hi Pos Hi Neg P-value

18 2

12 (67%) 1 (50%)

0.64

11 (61%) 2 (100%)

0.27


13 7

9 (69%) 4 (57%)

0.59

Control cohort


15 1

11 (73%) 1 (100%)

0.55

10 (67%) 1 (100%)

0.49


12 4

9 (75%) 2 (50%)

0.35

Number and percentage (in brackets) of samples in which bacteria were detected. p-value (bold when significant) as determined by Chi-Square analyses. Percentage calculated as percentage of the bacteria positive and bacteria negative value.

99

3.3.5 Bacterial load in the nasal, nasopharynx and adenoids of

children with and without RAOM/COME

The pathogen specific bacterial loads present within the nasopharynx and

adenoids were examined using samples from children with and without a history

of RAOM/COME using semi-quantitative culture and quantitative PCR (qPCR) to

identify potential differences.

In the nasopharynx and nasal vestibule, bacterial loads did not differ significantly

between children from the OM or Control cohorts (Figure 3-2).

Bacterial load estimates from the adenoid samples, analysed using semi-

quantitative culture, identified no significant difference between children with and

without RAOM/COME (Figure 3-2). In contrast, qPCR of the same demonstrated

significantly higher bacterial load of Mcat in the adenoids of children with OM,

compared to adenoid samples from control group children (p=0.001; Figure 3-2).

No significant differences in bacterial loads of other otopathogens were observed

in the adenoids between children with OM and control children (Figure 3-2).

100

a. Culture loadQ

ua

dra

nts

of

str

ea

kin

g

Spn H

i

Mca

t0

1

2

3

4

b. qPCR load

log

10

of

ba

cte

ria

ce

lls (

ml)

-1

Spn H

i

Mca

t0

2

4

6

8

a. Culture load

Qu

ad

ran

ts o

f str

ea

kin

g

Spn H

i

Mca

t0

1

2

3

4

b. qPCR load

log

10

of

ba

cte

ria

ce

lls (

ml)

-1

Spn H

i

Mca

t0

2

4

6

8

a. Culture load

Qu

ad

ran

ts o

f str

ea

kin

g

Spn H

i

Mca

t0

1

2

3

4

b. qPCR load

log

10

of

ba

cte

ria

ce

lls (

ml)

-1

Spn H

i

Mca

t0

2

4

6

8p=0.001

Figure 3-2: Bacterial loads were determined in the NS, NPS and adenoidal samples from children with and without RAOM/COME

using semi-quantitative bacterial culture and qPCR. Values in both graphs are mean SEM. Unpaired t-test was performed for

comparison.

OM2 cohort Control cohort

NS NPS Adenoid samples

101

3.3.6 Pneumococcal serotypes in the upper respiratory tract of


Spn isolates from NPS, NS and adenoids from the OM2 and control cohorts were

serotyped using Quellung reactions for the 23 serotypes/groups included in the

23-valent pneumococcal polysaccharide vaccine (PPV23). Serotyping of Spn

isolates from the OM1 cohort was not performed.

Five Spn isolates from the NPS, 4 from NS and 7 from the adenoids of children

with a history of RAOM/COME were typed and of these 16 isolates, 31% were

serogroup 23 (23A/23B/23F), with serotype 23F included in the PCV7 and

PCV13. In addition, 31% of Spn isolates were serogroup 11

(11A/11B/11C/1D/11F), which are non-PCV13 serotypes. A total of 38% of Spn

isolates were non-PPV23 vaccine groups/types (Table 3-6).

In comparison, five Spn isolates from the NPS, 3 from the NS and 7 from the

adenoid were tested from control group children. Of the 15 isolates, 47%, 13%,

7% and 38% were serogroup 23, 11, 19 (19A/19L/19N/19V) and non-PPV23

vaccine serotypes/groups, respectively (Table 3-6).

Overall, 58% (18/31) of pneumococcal serotypes isolated from children with and

without RAOM/COME, were non-PCV13 serotypes (Table 3-6). All Spn isolates

from the NPS and NS of the same child had the same serogroups/types. Only

62.5% of isolates from the adenoids had the same serogroups/types to those

isolated from NPS or NS of the same child.

102

Table 3-6: Pneumococcal serotypes in children with and without RAOM/COME.

OM cohort 2

(n = 16)

Control cohort

(n = 15)

Total

(n = 31)

Serogroup 23 5 (31%) 7 (47%) 12 (39%)

Serogroup 19 0 (0%) 1 (7%) 1 (3%)

Serogroup 11* 5 (31%) 2 (13%) 7 (23%)

Non-PPV23 vaccine serogroups*

6 (38%) 5 (33%) 11 (36%)

PPV23: 23-valent pneumococcal polysaccharides vaccine contains 23 pneumococcal serotypes, including 1, 2, 3, 4, 5, 6B, 7F, 8, 9N, 9V, 10A, 11A, 12F, 14, 15B, 17F, 18C, 19A, 19F, 20, 22F, 23F and 33F. *Bold serogroups are not included in PCV13.

3.3.7 Hi typing for samples from the upper respiratory tract of


Hi isolates from the OM2 and control cohorts were tested for the presence of

capsulation by PCR for the bexA gene. All strains were negative for the bexA

gene indicating they were non-capsular or non-typeable. Unfortunately, Hi

isolates from the OM1 cohort were not available for typing.

103

3.3.8 Antibiotic susceptibility of bacterial isolates from the upper

respiratory tract of children with and without RAOM/COME

3.3.8.1 S. pneumoniae

Six pneumococcal isolates from the NPS and 3 from the MEE from the

OM1 cohort were tested for antibiotic susceptibility and demonstrated that

2 isolates were resistant to erythromycin whilst 7 were susceptible to all

antibiotics tested.

In the OM2 and control cohorts respectively, a total of 33 pneumococcal

isolates from the NPS (5 and 5), NS (4 and 3) and adenoids (8 and 8) were

tested for antibiotic susceptibility. Of these 33 isolates, 8 were resistant to

erythromycin, 7 to oxacillin, 1 to cotrimoxazole and 1 to vancomycin. Multi-

drug resistance was observed in 6 of the 33 isolates, 4 of these isolates

were sampled from the adenoid, however, 23 isolates were susceptible to

all antibiotics tested. When isolates from different URT sites within the

same child were examined, the same patterns of antibiotic susceptibility

were observed.

3.3.8.2 H. influenzae

Within the OM1 cohort, 6 Hi isolates from NPS and 1 isolate from MEE

samples were tested for β-lactamase production and antibiotic

susceptibility, with 1 isolate positive for β-lactamase production and

resistant to ampicillin. Two isolates, negative for β-lactamase, were also

resistant to cotrimoxazole.

104

Within the OM2 and control cohorts, respectively, 32 Hi isolates from the

NPS (8 and 6), NS (2 and 2), adenoid (7 and 6) and MEE (1) samples were

assessed. Twenty isolates (66%) were positive for β-lactamase and

resistant to ampicillin. Within the total of 32 isolates, 13 were resistant to

cotrimoxazole, 4 to amoxicillin/clavulanic acid, 1 to ceftriaxone and 1 was

resistant to tetracycline. Multi-drug resistance was observed in 13 isolates,

8 of which were isolated from the adenoids.

Hi strains isolated from NPS and NS samples of the same child had the

same patterns of antibiotic susceptibility. Nine of the ten isolates from

adenoid samples had the same results of antibiotic susceptibility with those

isolates from NPS or NS samples of the same child.

3.3.8.3 M. catarrhalis

Within the OM1 cohort, all 4 Mcat isolates identified in the NPS were

positive for β-lactamase, resistant to ampicillin, and susceptible to

cotrimoxazole.

In the OM2 and control cohorts, respectively, 32 Mcat isolates from the

NPS (9 and 8), NS (6 and 2) and adenoids (5 and 2) tested positive for β-

lactamase and were resistant to ampicillin. Two isolates were resistant to

amoxicillin/clavulanic acid and 1 was resistant to cotrimoxazole, whilst

multi-drug resistance was observed in 3 isolates.

Mcat isolates from the NPS and NS of the same child demonstrated the

same patterns of antibiotic susceptibility, with 5 of 6 isolates from the

adenoids also sharing the same patterns of antibiotic susceptibility as those

from the NPS or NS of the same child.

105

3.3.9 Bacterial biofilms and intracellular bacteria in MEE of children

with RAOM/COME

MEE samples from children with RAOM/COME were examined using

LIVE/DEAD staining to examine the viability and morphological

organisation of.

LIVE/DEAD staining works on an exclusion principle where SYTO9 will

penetrate any membrane, including those which are intact, whilst

Propidium iodide (PI) will only penetrate compromised membranes, but the

dye out-competes the binding sites of SYTO9.

The resulting pattern is green staining for live bacteria and red staining for

dead bacteria. Only eighty three of the 85 MEE samples collected were

stained as 2 samples had insufficient volume available for the analysis.

Forty-six of the 83 stained MEE samples (55%) were positive for the

presence of any bacteria, with bacterial biofilms or microcolonies, the

smallest form of biofilms, identified in 21 positive samples (Figure 3-3a

and b). Intracellular bacteria, within middle ear epithelial cells, were

positively identified in 2 middle ear samples (Figure 3-4).

106

Figure 3-3: Bacterial microcolonies and biofilms identified in MEE

samples.

a

This is a MEE sample from right ear of child no. 16 of the OM2 cohort. The sample

was stained with SYTO9 for viability of live cells (green, the top left image) and

PI for dead cells (red, the top right image). The merged image (the bottom image)

of top left and right image showed microcolony of live bacteria (orange arrow) in

the MEE of the child with RAOM/COME. The scale bar is 10µm.

10µm 10µm

10µm

107

b

These images are representative of typical LIVE/DEAD staining of the MEE,

where SYTO9 stains for viability of live cells (green, the top left image) and PI

stains for dead cells (red, the top right image). A microcolony of live bacteria

(green) and dead bacteria (red) is highlighted within in a blue circle on the merged

image (the bottom image) of the MEE of child number 23 from the OM2 cohort,

that is a child undergoing treatment for RAOM/COME. Extracellular DNA is

illustrated by the white arrow and the scale bar is 10µm.

10µm 10µm

10µm

108

Figure 3-4: Representative image of LIVE/DEAD stained MEE from a child with

a history of RAOM/COME (OM2 Child 32 right ear) showing the complex 3-

dimensional (3-D) morphology of the bacterial biofilms and the nature of bacterial

involvement, live or dead, intracellular, and extracellular within the biofilm. Live

cells (green; SYTO9 stained), dead bacteria (red; PI stained), live intracellular

bacteria (yellow arrow) and live extracellular bacteria (white arrows) and dead

extracellular bacteria (blue arrows) are shown.

Width: 69.35µm; height: 69.35µm; Depth: 7.00µm

109

3.3.10 Bacterial biofilms on middle ear biopsies of children with

RAOM/COME

The presence and morphological organisation of the predominant

otopathogenic bacterial species in middle ear biopsies from children with

RAOM/COME, were examined using fluorescent in situ hybridisation

(FISH). Specific probes for Spn, Hi, Mcat and a universal bacterial probe

were used. A maximum of 3 probes, including the universal probe, could

be used on each biopsy as well as a nucleic acid stain (Hoechst 33342).

Eleven samples were hybridised with the specific probes to determine the

presence of Spn, Hi, Mcat and any bacteria. Eight samples (73%) were

positive for any bacteria and most of the bacteria detected were in

microcolony forms (Figure 3-5). Of 8 positive samples, 4 samples were

positive for Spn and Hi each, whilst only 1 sample was positive for Mcat.

Multiple-species of bacteria were co-identified in 2 samples, with Hi and

Spn present in one sample and Hi and Mcat in the other (Table 3-7).

Table 3-7: Summary culture and PCR results of 11 MEE samples and FISH

results of 11 MEM samples from children with RAOM/COME.

Child no.

Ear MEE culture results

MEE PCR results MEM FISH

Spn Hi Mcat Spn Hi Mcat EUB Spn Hi Mcat

01 L - - - - - - - - - na

01 R - - - - - - + + na -

04 L - - - - - + + - - na

06 L - - - - - - - - - na

07 R - - - - - - - - - na

11 L - - - - - - + na + +

11 R - - - - - - + na + -

20 R - - - + + - + + + na

21 L - - - - - - + + na -

21 R - - - - - - + na + -

22 L + - - + - - + + - na

Positive from number tested

1/11 (9%)

0 (0%)

0 (0%)

2/11 (18%)

1/11 (9%)

1/11 (9%)

8/11 (72%)

4/8 (50%)

4/9 (44%)

1/5 (20%)

EUB – universal (eubacterial) probe; Spn – S. pneumoniae (organism and probe); Hi – H. influenzae (organism and probe); Mcat – M. catarrhalis (organism and probe); na – not available.

110

Figure 3-5: Bacterial microcolonies identified in the middle ear biopsy from a child

with RAOM/COME (child no. 01, right ear). MEE samples of this child were PCR

and culture negative for all three main otopathogens Spn, Hi and Mcat. FISH was

performed on the ME biopsy using probes for Mcat (green), Spn (grey) and

universal bacteria (red). Hoechst staining for nuclei (blue) was also performed

and the specimen was imaged using CLSM. a) demonstrates Hoechst staining

of host cell nuclei; b) shows background auto-fluorescence since Mcat was not

identified in this sample; c) shows unidentified bacteria present within the sample

(white arrows; red); d) shows that Spn was identified in microcolony formation

and scattered throughout the biofilm (white arrows; grey); e) overlay of each

individual channel (images a, b, c and d) further highlights the micro-colonies of

Spn (white arrows) observed in this sample. The scale bars are 50µm.

a b

c d

e

50µm 50µm

50µm 50µm

50µm

111

3.4 Discussion

This study aimed to describe the bacterial aetiology of RAOM/COME in urban

children of South East Queensland. The frequency of identification of each of the

three predominant bacteria causal for OM were investigated in several URT

locations within the same child, including the middle ear and the potential

concurrent colonisation of the same bacteria was explored using antibiotic

sensitivity for these isolates as well as serotyping where relevant. The potential

impact of environmental factors including the breast feeding, pacifier use,

daycare attendance and immunisation history were explored as factors that may

guide future monitoring and prevention strategies.

Overall, regardless of their previous history for RAOM/COME, examination of

samples of the URT of urban children frequently identified the three main bacterial

otopathogens considered causal for OM, within multiple URT locations.

Detection frequencies for each predominant bacteria, Hi, Spn and Mcat were

highest in the adenoids followed by in the nasopharynx, nasal vestibule and then

the middle ear (in cases of children with RAOM/COME). These findings are

consistent with a recent report on microbiome of otitis media in Aboriginal

Australian children, who typically experience significantly increased risk of OM.

The report demonstrated that Streptococcus sp., H. influenzae and M. catarrhalis

were common operational taxonomic units (OUTs) found in the adenoids,

nasopharynx, and middle ear (Jervis-Bardy et al., 2015). Furthermore, the

relatively low detection frequencies of these bacteria in the nasal vestibule

compared with detection within the nasopharynx and adenoids was supportive of

previous research that hypothesised that nasal cavities are not the preferred

habitat for these bacteria, compared to S. aureus (Lemon et al., 2010, Wislson,

2005). Lower frequencies of Hi, Spn and Mcat detection in the nasal cavities

compared to the nasopharynx has also been reported for patients with viral

infection compared to a control group of healthy children (Rawlings et al., 2013).

The current study provided additional evidence of the potential association of

bacteria present within the nasal vestibule being concurrently detected in the

nasopharynx and adenoids within the same child, as illustrated from antibiotic

sensitivity and bacterial typing results. For example, all Hi and Mcat detected in

112

the nasal vestibule were also detected in the nasopharynx of the same children

and these isolates shared the same antibiotic sensitivities which supported co-

colonisation with the same bacteria. The similarity of these characteristics of the

bacteria and the frequency of their identification, despite the different

environments they occupied in the URT, support the hypothesis that they co-

colonised different locations in the URT and that, due to the higher frequency of

the identification within the nasopharynx and adenoids, these regions are more

likely to be reservoirs of bacterial otopathogens identified within the middle ear

compared to the more exposed, nasal vestibule.

Most children in this study, regardless of OM history, were colonised by at least

one of the 3 main otopathogens in the nasopharynx and/or adenoids, although

no significant differences were observed between children with or without a

history of RAOM/COME in the current study. Interestingly, a larger study

conducted in Western Australia (WA) reported that children significantly impacted

by RAOM at a younger age had significantly higher rates of nasopharyngeal

colonisation with Hi and Spn compared to healthy controls (Wiertsema et al.,

2011b). These younger children (mean age =1.7 years, n=186) were more

severely affected by RAOM than observed for the older (mean age =4.1 years,

n=43) children being treated for RAOM/COME in the current study. The younger

children were recruited to the previous study during surgical treatment

(tympanostomy tube insertion). The developmental progression of OM severity

and frequency generally reduces as the child’s face grows (Bluestone, 1996,

Swarts et al., 2013), which, in addition to the smaller cohort size, may help to

explain the lack of significance between the OM affected children and control

children of a similar age.

Selection of appropriate control groups remains an ongoing challenge for

investigations of young children. The control group selected for the current study

recruited children of similar age, in equivalent numbers of males and females,

who were undergoing surgery for AH/OSA. This control group has been

frequently used in past reports for comparison of OM and “healthy” control groups

(Hoa et al., 2010, Nistico et al., 2011, Almac et al., 2009, Saafan et al., 2013).

113

Results from the current study contributed further evidence that AH/OSA patients

may not best represent “healthy” control children. The concern that adenoid

hypertrophy, despite an absence of a clinical history of OM, may also have been

caused by chronic infection of the upper respiratory has been reported (Rout et

al., 2013).

Selection, recruitment and sampling from appropriately age and gender matched

children remains an ongoing challenge for similar studies since sample collection

from multiple locations in the URT is optimal for young children whilst they are

under anaesthesia. Control group children recruited to the WA study by

Wiertsema and colleagues (Wiertsema et al., 2011b) were predominantly male

children, since the majority of surgery undertaken in otherwise healthy children

within the age group was for the treatment of strabismus, hypospadias repair or

circumcision, which precludes adequate representation of female children with

OM. Another larger study, identified bacterial colonisation of the nasopharynx in

children with a history of RAOM/COME (n=198) compared to 55 healthy controls

(children undergoing minor surgeries in the absence of significant history of OM

or AH/OSA) using culture techniques. For these children, the frequency of Hi

detection in the nasopharynx of children with RAOM/COME was higher than

healthy controls (Marchisio et al., 2003). The average age of the children was 3.7

years for the RAOM/COME patients and 3.9 years for control children, which are

comparable to the current study.

Although the use of culture-based detection methods (in Marchisio’s study) may

have under-reported the frequency of bacterial detection, the older age of children

may be another factor reducing percentage of bacterial detection. The current

results confirm that use of culture-based techniques would result in under-

reporting of detection, but this issue would have been consistent between both

control and OM groups. Compared to the current study, it is clear that child age

and control group selection need careful consideration in future studies.

Co-detection of different bacteria in different locations of the URT in this study

has not only confirmed the role of the nasopharynx and adenoids as potential

sources of Hi and Mcat but has also demonstrated that Spn detection in the

middle ear of children with RAOM/COME is also linked to colonisation of the

114

nasopharynx and/or adenoids of the same child. Since all Spn isolated from the

MEE was also isolated from either the NPS or Adenoids then it is hypothesised

that Spn in the middle ear has originated from the nasopharynx and/or adenoids.

Although Hi and Mcat were also shown to originate from the nasopharynx and/or

adenoids in the current study, it is important to recognise that not all Hi or Mcat

were concurrently identified in the MEE, nasopharynx and/or adenoids in the

same child. A report by Stol et al., 2013, supports these findings having shown a

genetic match between Spn found in the middle ear and nasopharynx of children

with RAOM/COME. Importantly, these children were undergoing ventilation tube

surgery, similar to the current study and were not experiencing current active OM

infection. Furthermore, only 80% of Hi isolates from the middle ear and

nasopharynx were a genetic match (Stol et al., 2013). Together the current

findings and those by Stol, suggest that Hi or Mcat in the middle ears of children

with OM may persist and are not necessarily seeded from those bacteria in the

nasopharynx and/or adenoids.

Persistent and/or recurrent bacterial infections of the middle ear were identified

in the middle ear of children with RAOM/COME, who were otherwise well and

undergoing ventilation tube surgery in the current and previous studies. Bacteria,

located within bacterial biofilms and intracellularly within the epithelial cells of the

middle ear were visualised using FISH and CLSM in the current and previous

investigations (Hall-Stoodley et al., 2006, Thornton et al., 2011, Thornton et al.,

2013). Furthermore, 62% of NTHi detected in MEE at a recurrence of AOM were

genetically similar to NTHi detected in the initial AOM episode for the same child

(Kaur et al., 2013). Thus, during RAOM, up to two-thirds of NTHi may persist in

the middle ear from the initial infection, possibly through biofilms or intracellular

infection and contribute to reinfection of the middle ear.

In contrast to NTHi, Spn and Mcat were concurrently and significantly more

frequently detected in the nasal vestibule and nasopharynx of children with

RAOM/COME compared to control group children. These data suggest that co-

localisation of Spn and Mcat (rather than presence of either alone) might be

considered as a contributor to OM, however the numbers in the current study

were too small to further examine this finding. Supportive evidence from animal

115

models indicate that mice with nasal co-infection by Spn and Mcat resulted in

increased rates of pneumococcal OM and extended the duration of infection,

when compared to mice co-infected by Spn and Hi (Krishnamurthy et al., 2009).

Furthermore, the rate of successful ascension of Spn to the middle ear was

enhanced in animal models in association with Mcat co-infection (Perez et al.,

2014).

Investigation of the bacterial aetiology of OM in urban children in this study also

identified that the nasopharynx and nasal vestibule samples had similar bacterial

loads in children with RAOM/COME compared to controls. These results contrast

to a previous study, where high loads of otopathogenic bacteria in the

nasopharynx were correlated to AOM in Indigenous Australian children in remote

areas (Binks et al., 2011).

A significant complication in the interpretation of Binks’s study, arises from the

wide variation of the clinical morphology of the tympanic membrane observed in

the patients. Although NPS samples were collected from all children, 16% of the

336 AOM samples were collected from children with a TM perforation, 31% of

AOM samples were from patients without a TM perforation, whilst over the cohort,

48% of patients had OME and 5% were from a normal ear (Binks et al., 2011).

The extremely high bacterial loads and significantly increased risk of OM

experienced by Indigenous Australia children from remote areas results in very

young children experiencing very severe disease (Leach et al., 1994, Smith-

Vaughan et al., 2006). Indeed, OM, tympanic perforation and suppuration have

been reported from three weeks of age in these children (Smith-Vaughan et al.,

2008). Comparison of aetiological results from this cohort to the current study is

again difficult, since the average age of children in the study by Binks (2011) was

less than 2 years compared to over 4 years in the current study. Interestingly, the

current study identified increased bacterial load of Mcat in the adenoids of

children with OM compared to children from the Control group. Unfortunately, the

cohort size and detection frequency of Mcat precluded further examination of this

finding. Further research is needed to determine whether the bacterial load of

Mcat, either alone or in conjunction with other bacteria, may contribute to

development of RAOM/COME in urban children.

116

Throughout this study, Hi was the predominant otopathogen detected in the URT,

particularly in the nasopharynx and adenoids of the children with and without

RAOM/COME in addition to the middle ear of children with RAOM/COME. The

current results are consistent with reports from Western Australia (Wiertsema et

al., 2011b) and New Zealand (Mills et al., 2015) and the reflect the regional

prevalence of Hi as the most common bacterial otopathogen in the middle ear

and nasopharynx of children with RAOM or COME reported in a global

systematic review (Ngo et al., 2016) Importantly in this study, the predominant

otopathogen, Hi was progressively increasingly resistant to β-lactam antibiotics

rising from 14% (2008-2011) to 66% (2015) in the most recent cohort in the

current study. These findings highlight the importance of continuous surveillance

and progress toward vaccine development for this bacteria. Recently, PHiD-CV

was reported to reduce the frequency of middle ear infection caused by NTHi in

Australian Indigenous communities (Leach et al., 2015). Development of an

efficacious vaccine for Hi has potential benefits for reduction in OM prevalence in

both urban children and Indigenous Australian children, who are known to

experience significantly increased risk of severe OM development at a young age

(Leach et al., 1994, Boswell and Nienhuys, 1996). In the current study, although

Spn and Mcat were detected at similar or lower frequencies compared to Hi in

the different regions of the URT in children with and without RAOM/COME, these

results were consistent with the results of a systematic review of previous reports

from the Pacific region, including Australia and New Zealand (Ngo et al., 2016).

Further examination of the bacteriology of the middle ear with PCV7 or PCV13

used as a determinant, demonstrated that only the frequency of Mcat detection

from the middle ear was significantly higher in PCV13-vaccinated children

experiencing RAOM/COME. Slight but non-significant reduction in Spn detection

frequency was observed in the current study however further recruitment would

be necessary to further clarify this result.

This study provides the first report of potential changes in bacteria identified

within the middle ear of children with RAOM/COME since the inclusion of PCV13

into the Australian NIP in 2011. This preliminary finding could indicate that PCV13

may have an improved efficacy to reduce Spn colonisation in the middle ear,

117

which may also reduce OM in young children. Indeed, lower frequency of

detection for Spn isolations from MEE of children with AOM was observed in most

developed countries where PCV7 introduction has occurred (Casey et al., 2010,

Dupont et al., 2010, Hoberman et al., 2011, Alonso et al., 2013). The rising

frequencies of non-PCV7 serotypes, particularly serotype 19A has resulted in

Spn remaining, and in some countries, the predominant bacteria isolated from

children with AOM (Casey et al., 2010, Fenoll et al., 2011). Introduction and use

of PCV13, which includes serotype 19A, into the NIP of several developed

countries, including the US and Australia (Kaplan et al., 2013, Johnson et al.,

2012), has resulted in reduction of serotype 19A identification amongst

pneumococcal strains isolated from the middle ear of children in the US (Kaplan

et al., 2015). Concurrent to the reduction in serotype 19A, other non-PCV13

serotypes, including serotype 35B, 23A, 23B, 15, and 11 were isolated more

frequently (Kaplan et al., 2015).

In the current study, although not significantly reduced, the frequency of Spn

detected in MEE tended to fall after the introduction of PCV13. Concurrently,

58% of the Spn strains isolated from the URT of all children, with and without

RAOM/COME were non-PCV13 serotypes, which is consistent with recent

studies in other countries (Ben-Shimol et al., 2014, Martin et al., 2014a, Kaplan

et al., 2015, Ozawa et al., 2015, Angoulvant et al., 2015). Further exploration of

these results using larger cohort sizes is necessary to determine the extent to

which PCV13 may provide effective and long term reduction in the frequency of

pneumococcal infection, including OM, in the future.

In this study, children experiencing RAOM/COME who were immunised using

PCV13 showed higher frequencies of Mcat detection in their nasopharynx and

middle ear. Isolated detection of Mcat in the middle ear was increased 10-fold

subsequent to the use of PCV-13 in the NIP. In addition, Mcat detection in the

middle ears was increased three-fold, in children fully vaccinated using PCV-13.

The increasing predominance of Mcat as the predominant bacteria was reported

for children with RAOM, post-PCV-13 immunisation, in the USA (Casey et al.,

2015a). Together, the current results and the report by Casey and co-worker

(2015), suggest that Mcat may contribute more significantly to AOM after PCV13

118

immunisation, compared to PCV7 vaccination. Furthermore, Mcat infections

causal for AOM have been associated with lower frequencies of spontaneous

perforation of the tympanic membrane, but involving younger children at

diagnosis and increasingly involved in higher proportions of mixed bacterial

infections, compared with AOM caused by other bacterial otopathogens (Broides

et al., 2009). Co-infection of Mcat and Spn resulted in increased rates of

pneumococcal OM and extended the duration of infection in animal models

(Krishnamurthy et al., 2009, Perez et al., 2014). Currently, despite investigation

of a number of potential vaccine candidates, there remains no vaccines approved

for use against this bacteria.

In summary, the bacterial aetiology of RAOM/COME within the URT of urban

children from South-East Queensland included the three predominant pathogens,

Spn, Hi and Mcat, known to be causal for OM. Although Hi was the predominant

bacteria in the URT, it has become increasingly resistant to β-lactam antibiotics,

particularly since the introduction of PCV-13 in 2011. Although the nasopharynx

and adenoid play key roles in seeding otopathogenic bacteria to colonise the

middle ear, bacteria located within biofilms or epithelial cells of the middle ear

contribute significantly to recurrence or persistence of OM infection. This study

provided evidence of changes in environmental factors and bacterial detection

frequencies, that highlights the importance of continuous bacterial surveillance,

progressive vaccine development and informed treatment strategies for OM.

Improved treatment and potential prevention of OM has the potential to

significantly improve the quality of life for many children, world-wide, who

experience OM regularly during their first years of life and live with the life-long

damage associated with early hearing loss.

119

Chapter 4:

Viruses identified within the upper

respiratory tract of urban children with

and without RAOM/COME in South East

Queensland

120

4.1 Introduction

Viruses, particularly respiratory viruses are increasingly recognised for their

importance in the pathogenesis of OM. They are thought to act as a precursor

that facilitates concurrent or subsequent bacterial infection within the middle ear

(Fainstein et al., 1980, Chonmaitree, 2006).

Viral infections within the middle ear can increase the adherence of

otopathogenic bacteria to epithelial cells (Heikkinen and Chonmaitree, 2003) and

induce Eustachian tube (ET) dysfunction, which impedes clearance of middle ear

effusion (MEE) from the middle ear (Marom et al., 2012). Meanwhile, host innate

immune responses including virus-induced cytokine, inflammatory mediator and

mucus production, may enhance the risk of and contribute to the complex

pathogenesis of OM (Heikkinen and Chonmaitree, 2003, Nokso-Koivisto et al.,

2006). In the infant mouse OM model, influenza virus has been shown to

contribute to the development and persistence of pneumococcal OM via the

induction of neutrophil extracellular trap production in the middle ear space (Short

et al., 2011). Post-viral accumulation of MEE, posterior to the tympanic

membrane, maybe observed weeks to months after clearance of the original

infection and is described as chronic otitis media with effusion (COME) (Rovers

et al., 2004). Frequent episodes of RAOM and persistent COME are frequently

treated using insertion of ventilation tubes through the tympanic membrane.

Using non-molecular methodologies, such as viral culture, viruses have been

detected in the nasopharynx of 30-50% of children with AOM and in 10-20% of

MEE samples (Heikkinen and Chonmaitree, 2003, Rosenblut et al., 2001). The

frequency of detection for viruses detected in samples from patients with OM

have increased significantly using PCR techniques, with studies showing that

viruses were present in nasopharynx of 90-94% of children with OM (Heikkinen

and Chonmaitree, 2003, Wiertsema et al., 2011b) and in 30-96% of MEE samples

(Monobe et al., 2003, Ruohola et al., 2006, Wiertsema et al., 2011b). Data on

identification of the predominant virus/es detected within the middle ear of

children with OM is inconsistent as methodologies differ. In studies using non-

PCR methods, respiratory syncytial virus (RSV) was predominant (Heikkinen et

al., 1999, Patel et al., 2007, Yano et al., 2009b), while those studies using PCR

121

methodologies reported that human rhinovirus (HRV) was predominant

(Pitkäranta et al., 1998, Ruohola et al., 2006, Wiertsema et al., 2011b, Stol et al.,

2013).

Recently, newly discovered viruses, such as bocavirus (HboV), human

metapneumovirus (hMPV), coronaviruses and polyomaviruses, have been

detected in the nasopharynx and MEE specimens of children with OM,

suggesting that they may contribute to OM pathogenesis (Beder et al., 2009,

Wiertsema et al., 2011b, Lehtoranta et al., 2012). In addition, these viruses are

also frequently identified within the adenoids of children (Comar et al., 2011,

Günel et al., 2015), which may indicate that the adenoids could contribute as a

reservoir for these viruses in middle ear infection.

This study aimed to identify specific viruses within the upper respiratory tract,

including the middle ear, of children with and without OM, to investigate their

potential co-occurrence and co-localisation with the 3 main bacterial

otopathogens. Multiplex real-time PCR assays were performed to determine the

presence of nucleic acids of 17 viruses, including influenza A/B virus (IA/BV),

parainfluenza virus 1/2/3 (PIV1/2/3), adenovirus (ADV), RSV, HRV, hMPV, HKU-

1 coronaviruses (HKU-1), OC43 coronavirus (OC43), NL63 coronavirus (NL63),

229E coronavirus (229E), enterovirus (EV), HboV, KI polyomavirus (KI) and WU

polyomavirus (WU) in the nasopharynx and the adenoids of children with and

without RAOM/COME and in MEE samples from children with RAOM/COME.

4.2 Methods

Patient recruitment, sample collection and processing, viral nucleic acid

extraction and viral PCR were all performed as described in Chapter 2, sections

1, 2, and 4 respectively. For the pilot OM1 cohort, a panel of 10 viruses including

ADV, hMPV, IAV, IBV, RSV, HRV, PIV1, PIV2, PIV3 and WU were examined.

The OM2 and control cohorts, were investigated using an expanded panel of 17

viruses, including 7 viruses, HboV, 229E, HKU-1, NL63, OC43, EV and KI in

addition to the original 10.

Data were analysed using the SPSS for Windows, version 23, statistical analysis

package. Host and environmental risk factors were compared between children

122

with RAOM/COME and children without RAOM/COME using Pearson Chi-square

analyses (p-value asymptotic significant 2-sided) for categorical variables, such

as gender and vaccination status, and unpaired t tests for age and number

episodes of OM. Upper respiratory tract carriage of viruses and specific

combinations of co-occurring species were compared between OM and control

cohorts using Pearson Chi-square and Pearson correlation analyses, with

p<0.05 considered significant

123

4.3 Results


This research examined viral identifications in the upper respiratory tract of the

same cohorts of children previously described in Chapter 3, Section 3.1. In brief,

the three cohorts were the OM1 (pilot), OM2 that included recruited children with

RAOM/COME and the control cohort, children who were undergoing

adenoidectomy for AH/OSA. Descriptive demographic information for each

cohort of children included in this viral study is shown in Table 4-1. There were

no significant differences in environmental factors, such as antibiotic use, having

siblings and attending day-care, as well as host factor, such as age and gender,

between OM and control groups (Table 4-1).

124


cohort enrolled in this study.

OM1 OM2 Control p-value a

p-value b

Number 23 20 17

Mean age in year (range) 3.3 (1-7) 4.1 (2–8) 4.4 (3–8) 0.16 0.60

Male (%) 12 (52%) 11 (55%) 12 (71%) 0.71 0.51


>4c 7.1 (4-11) 0.9 (0-2) na <0.001

Attending day-care na 18/19 (95%)

13/16 (81%)

na 0.21

Antibiotic use in the last month

na 11/19 (58%)

7/16 (44%)

na 0.40

Having siblings na 17/19 (89%)

15/16 (94%)

na 0.65

A history of breastfeeding na 16/19 (84%)

14/16 (88%)

na 0.78

A history of pacifier use na 11/19 (58%)

7/16 (44%)

na 0.40

Allergy to any substance d na 4e/19 (21%)

4f/16 (25%)

na 0.78

Asthma na 2/19 (11%) 4/16 (25%)

na 0.26

Exposed to cigarette smoke

na 6/19 (32%) 4/16 (25%)

na 0.67


23 (100%)

9 (45%) 9 (53%) 0.00 0.63


0 (0%) 11 (55%) 8 (47%) 0.00 0.63

a OM1 vs OM2 b OM2 vs Control c The specific frequency of AOM episodes not available for OM1 cohort d Cow’s milk protein, egg, wheat, shellfish, house dust mite, pollens/trees, moulds, grasses, bee/insect venom, peanut, other nuts, cat/dog hair, others. e Allergic to penicillin (n=2), strawberries and kiwi fruits (n=1) and cow’s milk protein (n=1). f Allergic to penicillin (n=1), wheat (n=1), gluten (n=1) and cow’s milk protein (n=1).

na: not available


Pearson Chi-Square analyses were used categorical variables (Male, attending day-care, antibiotics use, siblings, breastfeeding, pacifier use, allergy, asthma cigarette smoke expose and vaccination)

125

4.3.2 Viral carriage

To describe the carriage of common upper respiratory viruses by children with

and without a history of RAOM/COME, viral detection, within each sampled

location of upper respiratory tract are described below.

4.3.2.1 Viral identification from the adenoids of children with and

without RAOM/COME

Viral detection within the adenoid tissues of children with and without

RAOM/COME confirmed the presence and identity of at least one virus in every

adenoid sample collected from the children in each of the 3 studied cohorts

(Table 4-2). Within the OM1 (pilot) cohort, 5 of the 10 viral species tested, were

detected in the adenoids. Within the OM2 and control cohorts, of the 17 viral

species tested, 14 and 12 were present, respectively (Table 4-2). There were no

significant differences in the frequency or species detected in the OM2 and its

control cohort, however comparison of the detection frequencies of ADV, hMPV,

HRV and PIV3 between the OM 2 and OM1 cohorts were significantly higher in

the OM2 group (p<0.05; Table 4-2).

Overall, HRV was the most commonly detected virus within the adenoids and

was detected in 53%, 85% and 69% of adenoids from children in the OM1, OM2

and control cohorts, respectively. Indeed, HRV was the most commonly detected

virus in adenoids from OM2 and control children and the second most common

virus in the OM1 cohort following WU (Table 4-2). The frequency of HRV

detection within the adenoids from the OM2 group was significantly higher than

that observed in the OM1 cohort (p=0.03; Table 4-2) but did not differ to the

control group (p=0.24; Table 4-2).

WU polyomavirus was detected in 58%, 55% and 31% of adenoid tissues from

the OM1, OM2 and control cohorts, respectively (Table 4-2). While WU was the

most commonly detected virus in the adenoids of the OM1 group, it was not

detected as frequently as other viruses within the OM2 and control cohorts (Table

4-2).

126

ADV was detected in 32%, 65% and 44% of adenoids from the OM1, OM2 and

control children, respectively (Table 4-2). A significantly higher frequency of ADV

detection was observed in the OM2 compared to the OM1 cohort (p=0.04; Table

4-2) however, there was no significant difference between the OM and control

cohorts (p>0.05; Table 4-2).

PIV3 was identified significantly more frequently in adenoids from the OM2 and

control children when compared to the OM1 cohort (p=0.003 for OM1 vs OM2,

p=0.01 for OM1 vs control; Table 4-2) and overall detections within the adenoids

from OM1, OM2 and control cohorts were 11%, 55% and 50%, respectively

(Table 4-2).

PIV2 was detected in 11%, 10% and 31% of the adenoids from the OM1, OM2

and control cohorts, respectively (Table 4-2), with no significant differences

between detection frequencies observed in the 3 cohorts (p>0.05; Table 4-2).

Detection frequencies of hMPV in the adenoids of children from the OM2 and

control cohorts were 35% and 25%, respectively, however this virus was not

identified in those adenoids from the OM1 cohort. Significantly higher rates of

hMPV detection were observed in the OM2 and control cohorts, compared with

that in the OM1 cohort (p<0.05; Table 4-2).

HboV, EV, OC43 and KI were detected with higher frequencies in adenoid

samples from the OM2 and control cohorts but they were not included in the panel

of viruses investigated in the adenoids of the OM1 cohort (the pilot cohort) (Table

4-2).

RSV, IAV, IBV, PIV1, NL63, HKU-1 and 229E were detected at either very low

frequencies or were not detected in the adenoids from children from the three

cohorts.

Detection of multiple viruses within the same sample was a frequent observation

in the adenoids from children of the OM2 and control cohorts compared with the

OM1 cohort (p<0.01; Table 4-2). In contrast, the frequency of detection of a single

virus only within the adenoids from the OM1 cohort was significantly higher than

that from either the OM2 or control cohorts (p<0.01; Table 4-2).

127

Table 4-2: Viral detection from the adenoids, nasopharynx (NPS), nasal (NS) and middle ear (MEE) of children with RAOM/COME (OM1

and OM2) and children without RAOM/COME (control).

Adenoids NPS NS MEE

OM1

(n=19)

OM2

(n=20)

Control

(n=16)

p-value a p-value b p-value c OM1

(n=23)

OM 2

(n=20)

Control

(n=17)

p-value a p-value b p-value c OM2

(n=20)

Control

(n=16)

p-value

OM1

(n=45)

OM2

(n=36)*

p-value

Children positive for at least one studied virus

19 (100%) 20 (100%) 16 (100%)

na na na 16 (70%)

13 (65%)

8 (47%) 0.75 0.27 0.15 8 (40%) 6 (38%) 0.88 20 (44%) 15 (42%) 0.80

Adenovirus 6 (32%) 13 (65%) 7 (44%) 0.04 0.20 0.46 2 (9%) 3 (15%) 0 (0%) 0.52 0.10 0.21 0 (0%) 0 (0%) na 3 (7%) 0 (0%) 0.11

Bocavirus na 13 (65%) 11 (69%) na 0.81 na na 0 (0%) 0 (0%) na na na 0 (0%) 0 (0%) na na 3 (8%) na

Coronavirus

229E

HKU-1

NL63

OC43

na

na

na

na

0 (0%)

0 (0%)

2 (10%)

4 (20%)

0 (0%)

0 (0%)

1 (6%)

7 (44%)

na

na

na

na

na

na

0.69

0.12

na

na

na

na

na

na

na

na

0 (0%)

0 (0%)

0 (0%)

3 (15%)

0 (0%)

0 (0%)

0 (0%)

2 (12%)

na

na

na

na

na

na

na

0.77

na

na

na

na

0 (0%)

0 (0%)

0 (0%)

1 (5%)

0 (0%)

0 (0%)

0 (0%)

2 (13%)

na

na

na

0.42

na

na

na

na

0 (0%)

0 (0%)

0 (0%)

1 (3%)

na

na

na

na

Enterovirus na 14 (70%) 10 (63%) na 0.64 na na 3 (15%) 2 (12%) na 0.77 na 0 (0%) 0 (0%) na na 0 (0%) na

Human metapneumovirus

0 (0%) 7 (35%) 4 (25%) 0.004 0.52 0.02 3 (13%) 0 (0%) 0 (0%) 0.094 na 0.12 0 (0%) 0 (0%) na 2 (4%) 1 (3%) 0.69

Influenza virus

IAV

IBV

0 (0%)

0 (0%)

1 (5%)

0 (0%)

0 (0%)

0 (0%)

0.32

na

0.36

na

na

na

1 (4%)

0 (0%)

0 (0%)

1 (5%)

0 (0%)

0 (0%)

0.35

0.28

na

0.35

0.38

na

0 (0%)

0 (0%)

0 (0%)

0 (0%)

na

na

0 (0%)

0 (0%)

0 (0%)

1 (3%)

na

0.26

Respiratory syncytial virus

0 (0%) 1 (5%) 0 (0%) 0.32 0.37 na 1 (4%) 1 (5%) 0 (0%) 0.92 0.35 0.38 0 (0%) 0 (0%) na 3 (7%) 3 (8%) 0.78

Rhinovirus 10 (53%) 17 (85%) 11 (69%) 0.03 0.24 0.33 7 (30%) 9 (45%) 7 (41%) 0.32 0.82 0.48 6 (30%) 4 (25%) 0.74 10 (22%) 6 (17%) 0.53

Parainfluenza virus

PIV1

PIV2

PIV3

0 (0%)

2 (11%)

2 (11%)

3 (15%)

2 (10%)

11 (55%)

1 (6%)

5 (31%)

8 (50%)

0.08

0.96

0.003

0.41

0.11

0.77

0.27

0.13

0.01

0 (0%)

4 (17%)

0 (0%)

0 (0%)

0 (0%)

1 (5%)

0 (0%)

1 (6%)

0 (0%)

na

0.05

0.28

na

0.27

0.35

na

0.28

na

0 (0%)

1 (5%)

0 (0%)

0 (0%)

0 (0%)

0 (0%)

na

0.36

na

0 (0%)

0 (0%)

0 (0%)

0 (0%)

0 (0%)

0 (0%)

na

na

na

Polyomavirus

WU

KI

11 (58%)

na

11 (55%)

3 (15%)

5 (31%)

2 (13%)

0.86

na

0.15

0.83

0115

na

7 (30%)

na

1 (5%)

1 (5%)

0 (0%)

2 (12%)

0.03

na

0.35

0.45

0.01

na

0 (0%)

0 (0%)

0 (0%)

0 (0%)

na

na

4 (9%)

na

1 (3%)

0 (0%)

0.26

na

Single virus 12 (63%) 0 (0%) 1 (6%) 0.000 0.26 0.001 9 (39%) 5 (25%) 4 (24%) 0.32 0.92 0.30 8 (40%) 6 (38%) 0.88 19 (42%) 14 (39%) 0.76

Multiple viruses 7 (37%) 20 (100%) 15 (94%) 0.000 0.26 0.001 7 (30%) 8 (40%) 4 (24%) 0.75 0.29 0.63 0 (0%) 0 (0%) na 1 (2%) 1 (3%) 0.87

*Only 36 of 40 MEE samples were tested for viruses due to one sample had deficient volume for viral detection and 3 samples had inhibitors. a OM1 vs OM2; b OM2 vs control; c OM1 vs control; Pearson Chi-square analysis was used to compare frequencies of detection in each of the specimens; na: not available

128

4.3.2.2 Viral identification from the nasopharynx of children with and

without RAOM/COME

For the NPS, at least one virus was detected in 70%, 65% and 47% of children

of the OM1, OM2 and control cohorts, respectively (Table 4-2). Seven of 10 and

9 of 17 viral species were detected in the nasopharynx of children from the OM1

and OM2 cohort, respectively, whilst only 5 of 17 viral species were detected in

the NPS samples from the control cohort (Table 4-2). No significant differences

in detection of specific viruses, except WU, were observed in the nasopharynx of

children with and without RAOM/COME (Table 4-2).

HRV was the predominant virus identified in the nasopharynx from all 3 cohorts

and was detected in 30%, 45% and 41% of the nasopharynges of children from

the OM1, OM2 and control cohorts, respectively (Table 4-2)..

WU was detected in 30% of NPS samples from the OM1 cohort, which was similar

to the frequency of HRV detection from the same cohort. In contrast, WU was

only detected in 5% of NPS samples from the OM2 cohort and not at all from the

control cohort (Table 4-2). The differences in detection rates of WU were

significant between the OM1 and OM2 cohorts (p=0.03; Table 4-2) and the OM1

and control cohorts (p=0.01; Table 4-2). No significant difference was observed

in WU detection between the OM2 and control cohorts.

PIV2 detection within the NPS of the OM1 and control cohorts was similar, 17%

and 6% respectively, but this virus was not detected from the OM2 cohort (Table

4-2).

ADV, RSV, hMPV, IAV, IBV and PIV3 viruses were only detected at low

frequencies in NPS samples from both OM cohorts and were not detected within

NPS from the control cohort (Table 4-2).

HboV, PIV1, 229E, HKU-1 and NL63 were not detected in NPS samples from any

of the cohorts, where they were tested (Table 4-2), however EV and OC43

detection occurred at similar frequencies in the NPS of the OM2 (15%) and

control (12%) cohorts. Samples from the OM1 cohort were not assessed for these

viruses (Table 4-2).

129

Viruses were co-detected in 30%, 40% and 24% of NPS samples from the OM1,

OM2 and control cohorts, respectively (Table 4-2) with no significant differences

observed between groups. Similarity, the detection frequency of only a single

virus in the NPS did not differ between the three groups Table 4-2). Comparison

of the detection frequencies for at least one virus within the NPS and adenoids

showed that overall, the adenoids were significantly more likely to contain at least

one tested virus, with detection in adenoids being 100% for all groups (70% vs

100% in the OM1, p=0.008; 65% vs 100% in the OM2, p=0.004; 47% vs 100% in

the control, p=0.001).

4.3.2.3 Viral identification from the nasal vestibule of children with and

without RAOM/COME

For the OM2 and control cohorts, nasal swabs were collected from the nasal

vestibule and analysed for the presence of viruses using the 17 virus panel.

No significant differences were observed between the frequency of detections

between the two cohorts, with at least one tested virus being detected in OM2

cohort (40%) compared to the control cohort (38%, p=0.88; Table 4-2). Three

viral species, including HRV, PIV2 and OC43, were detected in the NS from OM2

cohort children, whilst only HRV and OC43 were identified within control cohort

children (Table 4-2). HRV was the most frequently detected virus in NS samples

from the OM2 (30%) and control children (25%) (Table 4-2).

Coronavirus OC43 detection frequency was low in the OM2 (5%) and control

cohorts (13%). PIV2 was only detected in the NS from OM2 cohort children (5%)

and not at all in the control group. No other viruses from the 17 tested were

identified in NS samples from these cohorts (Table 4-2). Importantly, all viruses

detected in the nasal vestibule from both OM2 and control cohorts were detected

as single virus (Table 4-2).

Similarly to the comparison of viral detection frequencies in the NPS and

adenoids, the NS samples had significantly lower rates of viral detection than the

adenoids (40% vs 100% in the OM2, p=0.000; 38% vs 100% in the control,

p=0.000). The slightly lower frequency of viral detection in NS samples compared

130

to the detections within the NPS was not significant (40% vs 65% in the OM2

cohort, p=0.11; 38% vs 47% in controls, p=0.58).

4.3.2.4 Viral detection in the middle ear of children with RAOM/COME

Forty five MEE samples from the OM1 cohort and 36 MEE samples from the OM2

cohort were examined for 10 and 17 viruses, respectively, using real-time PCR.

At least one virus was detected in 44% and 42% of MEE samples from the OM1

and OM2 cohort, respectively (Table 4-2). Five of 10 viral species were detected

within OM1 cohort MEE samples whilst seven of 17 viral species were identified

in the MEE samples from OM2 patients.

HRV, RSV, hMPV and WU were detected in the MEE from both OM cohorts

(Table 4-2). ADV was only detected in the OM1 cohort and IBV was only detected

in the OM2 cohort (Table 4-2). HboV and OC43, examined only within the OM2

patients, were detected in the MEE (Table 4-2).

HRV was the most common virus detected in the middle ear of children from both

OM cohorts and was detected in 22% (10/45) and 17% (6/36) of MEE samples

from the OM1 and OM2 cohorts, respectively (Table 4-2). RSV was detected at

low frequencies in the MEE from both OM cohorts (7% (3/45) OM1 versus 8%

(3/36) OM2, respectively) (Table 4-2). Similarly, WU polyomavirus was detected

at low frequencies in both the OM1 and OM2 cohorts ( 9% (4/45) versus 3%

(1/36), respectively (Table 4-2). hMPV was the only other virus detected in MEE

from children of the OM1 and OM2 cohorts and occurred in 4% (2/45) and 3%

(1/36) of patients in these groups, respectively (Table 4-2). Other tested viruses,

such as IAV and parainfluenza viruses, were not detected in MEE samples from

either of the OM cohorts (Table 4-2).

Most MEE viral detection occurred as a single virus (19/20 positive samples in

the OM1 cohort, 14/15 positive samples in the OM2 cohort) whereas multiple viral

detection was observed in only one MEE sample per OM cohort (Table 4-2).

Overall, virus was detected less often in MEE samples when compared to the

different URT locations sampled, particularly the adenoids. Viral detection from

the MEE compared to adenoids was significantly lower (p=0.000 in both OM

131

cohorts) whilst NPS detections were reduced but not significantly so (44% vs

70%, p=0.05 in the OM1 cohort; 42% vs 65%, p=0.09 in the OM2 cohort). Viruses

detected within each ear of the same child were assessed to determine whether

the same viruses were present. A total of 40 pairs of MEE samples (n=22 OM1

and n=18 OM2 cohort) from 40 children with RAOM/COME were analysed.

Twenty four pairs (60%) of these 40 paired samples were positive for virus in one

or both ears. Eight of the 24 pairs (33%) had the same virus identified in each ear

while 16 pairs (67%) had different viruses detected from left and right ears of the

same child (Table 4-3).

132

Table 4-3: Viral results in left and right ears of the same child from 24 children in

the OM1 and OM2 cohorts.

Cohort Patient Ear HRV IBV hMPV RSV ADV WU HboV OC43

OM1 Child No. 2

Left - - - - - - na na Right - - - - + - na na

OM1 Child No. 4

Left + - - - - - na na Right + - - - - - na na

OM1 Child No. 5

Left - - - + - - na na Right - - - - - - na na

OM1 Child No. 8

Left - - - - - + na na Right - - - - - - na na

OM1 Child No. 12


OM1 Child No. 13

Left - - - - - + na na Right - - - - - + na na

OM1 Child No. 14

Left + - - - - - na na Right - - - - - - na na

OM1 Child No. 15

Left - - - - + - na na Right - - - - - - na na

OM1 Child No. 16

Left - - + - - - na na Right - - + - - - na na

OM1 Child No. 18

Left + - - - - - na na Right - - - - - - na na

OM1 Child No. 19


OM1 Child No. 21


OM1 Child No. 23

Left - - - - - - na na Right - - - + - - na na

OM2 Child No. 6

Left - - + - - - - - Right - - - - - - - -

OM2 Child No. 7

Left - - - - - - - - Right - - - + - - - -

OM2 Child No. 8

Left - - - + - - - - Right - - - - - - - +

OM2 Child No. 12

Left + - - - - - - - Right + - - - - - - -

OM2 Child No. 14

Left - - - - - - + - Right - - - - - - - -

OM2 Child No. 23

Left - - - - - - + - Right - - - - - - + -

OM2 Child No. 24

Left + - - - - - - - Right - - - - - - - -

OM2 Child No. 26

Left - - - - - - - - Right - + - - - - - -

OM2 Child No. 31

Left - - - + - - - - Right - - - - - - - -

OM2 Child No. 35

Left + - - - - + - - Right + - - - - - - -

OM2 Child No. 37

Left + - - - - - - - Right - - - - - - - -

HRV: rhinovirus; IBV: influenza B virus; hMPV: human metapneumovirus: RSV: respiratory syncytial virus; ADV: adenovirus; WU: WU polyomavirus; HboV: bocavirus; OC43: OC43 coronavirus.

na is not available: “+” is detected; “-“ not detected

133

4.3.3 Co-detection of specific viruses in different locations of the

upper respiratory tract of the same child from the OM and

control cohorts

Viruses detected from the NS were compared with those from the NPS. HRV,

OC34 and PIV2 were all detected in the NS. All 10 HRVs detected in the NS (n=6

OM2; n= 4, control) were also detected in the NPS of the same child (Figure 4-

1), as were all 3 detections of OC43 (n=1, OM2; and n=2, controls) (Figure 4-1).

The sole detection of PIV2 in the NS from a child in the OM2 cohort was not

identified in the NPS of the same child (Figure 4-1).

Comparison of viruses detected between the nasopharynx and adenoids of the

same child showed that all 5 ADV identifications from the NPS were also

identified in the adenoids of the same child (n=2 OM1; n=3 OM2) (Figure 4-1).

Similarly, all OC43 (5), EV (5), PIV3 (1) and KI (1), viruses detected in the NPS

were also detected in the adenoids of the same children from the OM and control

cohorts (Figure 4-1). In contrast, only 78% of HRV (18/23) and 50% of WU (4/8)

detected in the nasopharynx were also detected in the adenoids of the same

children (Figure 4-1).

Viral detections within MEE were compared to those from the nasopharynx or

adenoids of the same children in the OM cohorts. Data from both ears (left and

right) of the same child were considered as one sample in this analysis. Thirty

seven pairs of MEE and NPS/adenoidal samples were assessed and all 10 HRVs

identified within the MEE were also detected in the NPS of the same children

(Figure 4-1). Interestingly, 70% (7/10) of HRV detected in the MEE were also

identified within the adenoids of the same child (Figure 4-1). Lower detection

frequencies of IBV and OC43 (n=1 each) in MEE were also detected in the

nasopharynx (Figure 4-1), whilst only 1 of the 2 ADVs detected in the MEE were

also detected in the nasopharynx and adenoids (Figure 4-1). RSV detections in

the MEE were detected in 50% (2/4) of NPS and none of the adenoids (Figure

4-1). One third (1/3) of WU detected in the MEE was also detected in the NPS or

adenoids (Figure 4-1) whilst none or 50% (1/2) of HboV detections in the MEE

were co-identified in the NPS and adenoids, respectively (Figure 4-1). hMPV

134

detections in the MEE were not identified in either the NPS or adenoids of the

same child (Figure 4-1).

ADV (100%, 5) EV (100%, 5) KI (100%, 3) OC43 (100%, 5) PIV3 (100%, 1) HRV (78%, 23) WU (50%, 8) hMPV (0%, 3) IAV (0%, 1) IBV (0%, 1) PIV2 (0%, 5)

OC43 (100%, 3) HRV (100%, 10) PIV2 (0%, 1)

HRV (50%, 4) OC43 (0%, 1)

OC43 (100%, 1) HRV (70%, 10) ADV (50%, 2) HboV (50%, 2) WU (33%, 3) hMPV (0%, 2) RSV (0%, 4)

IBV (100%, 1) OC43 (100%, 1) HRV (100%, 10) ADV (50%, 2) RSV (50, 4) WU (33%, 3) hMPV (0%, 2) HboV (0%, 2)

NPS

NS

MEE

Adenoids

HRV

ADV OC43

hMPV

IAV IBV

RSV

PIV2

PIV3 WU

KI

HRV

OC43

PIV2

HRV

WU HboV

ADV RSV

IBV

OC43

ADV HboV

NL63 OC43

EV hMPV

IAV HRV PIV2

PIV1 PIV3

WU

KI

hMPV

135

Figure 4-1: Co-detection of specific viruses in different locations of the upper

respiratory tract from children with and without RAOM/COME.

This figure represents the viruses detected using PCR in the different upper

respiratory tract (URT) locations for all children, with and without a history of

RAOM/COME and the associated locations where the same viruses were

identified in the same child.

Each circle represents an URT location and for each of the three bacteria, the

percentage of total viruses identified in the location and the number of children

are shown (x%, y). The size of the circles illustrating each URT location are

representative of the total number of viral detections (ex. NPS circle is larger than

NS; more viral detections in NPS).

Association of viral detections in one location compared to another are illustrated

using dash arrows. In children where the viruses detected in one area were also

detected in another area, this is demonstrated in the figure. For example, bold

red HRV (100%, 10) indicates that all HRV detected in middle ear were also

detected in NPS of the same child and 10 MEE samples positive for RHV were

compared to RHV detection results in 10 matching NPS samples from the same

children. Blue dash line indicates data from the OM2 and control cohorts; Green

dash line indicates data from the OM2 cohort only; Orange dash line indicates

data from the OM1 and OM2 cohorts. Red dash line indicates data from 3 cohorts

OM1, OM2 and control.

136

4.3.4 Co-detection of common viruses in the nasopharynx and

adenoids of children with and without RAOM/COME

Multiple viruses were co-detected frequently within the NPS (mean= 31% of NPS

samples; Table 4-2) and adenoids (mean=77% of adenoidal samples; Table 4-

2).

NPS samples from children with OM, demonstrated co-occurrences of WU

polyomavirus with either hMPV or ADV (Table 4-4). HRV and EV were also co-

detected within the NPS of children with OM (Table 4-4). No significant co-

occurrence of any viruses tested were observed in NPS of children from the

control cohort (Table 4-4).

Within the adenoid, identification of HRV was associated with the presence of

either PIV3 or HboV in children from both OM cohorts (Table 4-5). The presence

of 3 viruses, including hMPV, PIV3 and ADV, were positively correlated in the

adenoids of children with RAOM/COME (Table 4-5). In addition, the presence of

hMPV was positively related to the presence of OC43, and WU was positively

related to the presence of EV in the adenoids of children from the OM cohorts

(Table 4-5). In contrast, in adenoids of children from the control cohort, positive

correlations were observed between presence of hMPV and PIV2, as well as

between OCV43 and EV. Negative correlation were observed between the

detection of HRV and KI, as well as between OC43 and KI (Table 4-5).

Overall, adenoid samples from children with RAOM/COME show all positive

correlation in the presence of tested viruses, if occurred. In contrast, mixed

negative and positive correlations were observed in the presence of viruses in

the adenoids in children without RAOM/COME.

137

Table 4-4: Co-detection of common viruses in the nasopharynx of children from

the OM and control cohorts.

N hMPV PIV2 RSV ADV WU EV OC43 KI

Common viruses found in the nasopharynx from both OM cohorts

HRV Pos HRV Neg p-value

16 27

0 (0%) 3 (11%) 0.17

0 (0%) 4 (15%) 0.11

1 (6%) 1 (4%) 0.70

0 (0%) 5 (19%) 0.07

1 (6%) 7 (26%) 0.11

hMPV Pos hMPV Neg p-value

3 40

1 (33%) 3 (8%) 0.14

0 (0%) 2 (5%) 0.69

1 (33%) 4 (10%) 0.22

2 (67%) 6 (16%) 0.03

PIV2 Pos PIV2 Neg p-value

4 39

0 (0%) 2 (5%) 0.64

0 (0%) 5 (13%) 0.45

2 (50%) 6 (15%) 0.09

RSV Pos RSV Neg p-value

2 41

0 (0%) 5 (12%) 0.60

1 (50%) 7 (17%) 0.24

ADV Pos ADV Neg p-value

5 38

3 (60%) 5 (13%) 0.01

Common viruses found in the nasopharynx from the OM2 cohort


9 11

3 (33%) 0 (0%) 0.04

2 (22%) 1 (9%) 0.41


3 17

0 (0%) 3 (18%) 0.43

1 (33%) 2 (12%) 0.34

EV Pos EV Neg p-value

3 17

1 (33%) 2 (12%) 0.34

Common viruses found in the nasopharynx from the control cohort


7 10

2 (28%) 0 (0%) 0.07

2 (28%) 0 (0%) 0.07

1 (14%) 1 (10%) 0.77


2 15

1 (50%) 1 (7%) 0.07

0 (0%) 2 (14%) 0.58

OC43 Pos OC43 Neg p-value

2 15

0 (0%) 2 (14%) 0.58

Number and percentage (between brackets) of samples in which viruses were detected. p-value (in bold and red when significant) as determined by Chi-Square analyses. Percentage calculated as percentage of the bacteria positive and bacteria negative value.

138

Table 4-5: Co-detection of common viruses in the adenoids of children from the OM and control cohorts.

N hMPV PIV1 PIV2 PIV3 ADV WU KI OC43 EV HboV

Common viruses found in the adenoids from both OM cohorts HRV Pos HRV Neg p-value

27 12

6 (22%) 1 (8%) 0.30

3 (11%) 0 (0%) 0.23

4 (15%) 0 (0%) 0.16

12 (44%) 1 (8%) 0.03

14 (52%) 5 (42%) 0.557

14 (52%) 8 (67%) 0.39

hPMV Pos hPMV Neg p-value

7 32

1 (14%) 2 (6%) 0.47

1 (14%) 3 (9%) 0.70

5 (71%) 8 (25%) 0.02

6 (86%) 13 (41%) 0.002

3 (43%) 19 (59%) 0.43


3 36

1 (33%) 3 (8%) 0.17

2 (67%) 11 (31%) 0.20

1 (33%) 18 (50%) 0.58

2 (67%) 20 (56%) 0.71


4 35

2 (50%) 2 (6%) 0.46

2 (50%) 17 (49%) 0.96

1 (25%) 21 (60%) 0.18


13 26

11 (85%) 8 (31%) 0.002

8 (62%) 14 (54%) 0.65


19 20

9 (43%) 13 (65%) 0.27

Common viruses found in the adenoids from the OM2 cohort


17 3

3 (18%) 0 (0%) 0.43

4 (24%) 0 (0%) 0.35

13 (76%) 1 (33%) 0.13

13 (76%) 0 (0%) 0.01


7 13

0 (0%) 3 (23%) 0.17

4 (57%) 0 (0%) 0.002

4 (57%) 10 (77%) 0.36

5 (71%) 8 (62%) 0.66


11 9

2 (18%) 1 (11%) 0.66

3 (27%) 1 (11%) 0.37

9 (82%) 5 (56%) 0.20

8 (73%) 5 (56%) 0.42


13 7

1 (8%) 2 (29%) 0.21

4 (32%) 0 (0%) 0.10

9 (69%) 5 (71%) 0.92

8 (62%) 5 (71%) 0.66

WU Pos WU Neg p-value

11 9

3 (27%) 0 (0%) 0.09

1 (9%) 3 (33%) 0.18

10 (91%) 4 (44%) 0.02

9 (82%) 4 (44%) 0.08

KI Pos KI Neg p-value

3 17

0 (0%) 4 (24%) 0.35

3 (100%) 11 (65%) 0.22

3 (100%) 10 (59%) 0.17


4 16

2 (50%) 12 (75%) 0.33

3 (75%) 10 (63%) 0.64


14 6

11 (79%) 2 (33%) 0.05

139

Common viruses found in the adenoids from the control cohort


11 5

3 (27%) 1 (20%) 0.76

4 (36%) 1 (20%) 0.51

6 (54%) 2 (40%) 0.11

6 (54%) 1 (20%) 0.20

4 (36%) 1 (20%) 0.51

0 (0%) 2 (40%) 0.03

5 (45%) 2 (40%) 0.84

7 (64%) 3 (60%) 0.89

9 (82%) 2 (40%) 0.09


4 12

3 (75%) 2 (17%) 0.03

2 (50%) 6 (50%) 1.00

2 (50%) 5 (42%) 0.77

2 (50%) 3 (25%) 0.35

0 (0%) 2 (17%) 0.38

1 (25%) 6 (50%) 0.38

2 (50%) 8 (67%) 0.55

2 (50%) 9 (75%) 0.35


5 11

3 (60%) 5 (45%) 0.59

2 (40%) 5 (45%) 0.84

2 (40%) 3 (27%) 0.61

0 (0%) 2 (18%) 0.31

0 (0%) 7 (64%) 0.02

2 (40%) 8 (73%) 0.21

3 (60%) 8 (72%) 0.61


8 8

4 (50%) 3 (38%) 0.61

3 (38%) 2 (25%) 0.11

1 (13%) 1 (13%) 1.00

4 (50%) 3 (38%) 0.61

5 (63%) 5 (63%) 1.00

6 (75%) 5 (63%) 0.59


7 9

2 (29%) 3 (33%) 0.84

1 (14%) 1 (11%) 0.85

5 (71%) 2 (22%) 0.05

6 (86%) 4 (44%) 0.09

6 (86%) 5 (56%) 0.20


5 11

0 (0%) 2 (18%) 0.31

2 (40%) 5 (45%) 0.84

2 (40%) 8(73%) 0.21

4 (80%) 7 (64%) 0.51


2 14

1 (50%) 6 (43%) 0.85

1 (50%) 9 (64%) 0.70

1 (50%) 10 (71%) 0.54


7 9

7 (100%) 3 (33%) 0.006

6 (86%) 5 (56%) 0.20


10 6

7 (70%) 4 (67%) 0.89

Number and percentage (between brackets) of samples in which viruses were detected. p-value (in bold and red when significant) as determined by Chi-Square analyses. Percentage calculated as percentage of the bacteria positive and bacteria negative value.

140

4.3.5 Co-detection of common viruses and bacterial otopathogens

in the upper respiratory tract of children with and without

RAOM/COME

Bacterial and viral identifications within the URT of patients in this study (Chapter

3 and current chapter) were combined and examined in this section to provide an

overview of co-incident detection/infection of the URT locations sampled within

the patient cohorts of children with and without RAOM/COME.

Co-detection of bacteria and viruses in the adenoids were observed from all

children regardless of their history of RAOM/COME (Table 4-6). Overall, the

frequencies where no pathogen, bacteria alone and virus alone were detected

within the NPS were low, however these did not differ significantly between

children with OM and controls (5% vs 6%, 26% vs 47%, 9% vs 12%, respectively)

(Table 4-6). In contrast, children with RAOM/COME had higher frequency of co-

detection of viruses and bacteria in the nasopharynx than that in children of the

control group, however the difference was not statistically significant (61% vs

35%, p=0.09; Table 4-6).

The majority of MEE samples (67%, 54/81 ears) from both OM1 and OM2 cohorts

were positive for at least one virus or bacterium, with the remaining 33% of MEE

samples being negative (Table 4-6). Viral detection alone (without bacteria) was

observed in 18% of MEE samples whilst concurrent detection of both viruses and

bacteria were observed in 24% of MEE samples (Table 4-6).

141

Table 4-6: Microbial results in the upper respiratory tract of children from the OM

and control cohorts.

Adenoids OM2 cohort (n=20) Control cohort (n=16) p-value

No pathogen 0% 0% na

Bacteria alone 0% 0% na

Viruses alone 0% 0% na

Bacteria and viruses 100% 100% na

NPS OM cohorts (n=43)a Control cohort (n=17)

No pathogen 5% 6% 0.84

Bacteria alone 26% 47% 0.11

Viruses alone 9% 12% 0.77

Bacteria and viruses 61% 35% 0.09

MEE OM cohorts (n=81)a Control cohort (n=0)

No pathogen 33% na na

Bacteria alone 25% na na

Viruses alone 18% na na

Bacteria and viruses 24% na na

Pearson Chi-square analysis was used to compare between groups; na: not available a OM1 and OM2 cohorts.

142

The frequency of viral detection and concurrent detection of any of the three

predominant bacterial otopathogens (Chapter 3) was examined in each sampled

location within the URT of children with or without RAOM/COME. Only 6 viruses

were detected in more than one of the 81 MEE samples from children with

RAOM/COME in the OM1 and OM2 cohorts and concurrently detected with any

of the three predominant otopathogenic bacteria, Spn, Hi and Mcat. Only ADV

detection was weakly but significantly correlated with the presence of Hi in the

MEE (r=0.311, p=0.005; Table 4-7).


middle ear of children with RAOM/COME.

N Spn Hi Mcat

HRV Pos

HRV Neg

p-value

16

65

2 (13%)

11 (17%)

0.67

3 (19%)

19 (29%)

0.34

4 (25%)

13 (20%)

0.66

RSV Pos

RSV Neg

p-value

6

75

1 (17%)

12 (16%)

0.97

3 (50%)

19 (25%)

0.22

2 (34%)

15 (20%)

0.44

ADV Pos

ADV Neg

p-value

3

78

1 (33%)

12 (15%)

0.41

3 (100%)

19 (24%)

0.005

0 (0%)

17 (22%)

0.36

hMPV Pos

hMPV Neg

p-value

3

78

0 (0%)

13 (17%)

0.44

1 (33%)

22 (28%)

0.85

0 (0%)

17 (22%)

0.36

WU Pos

WU Neg

p-value

5

76

1 (20%)

12 (16%)

0.80

1 (20%)

22 (29%)

0.67

2 (40%)

15 (20%)

0.28

HboV* Pos

HboV Neg

p-value

3

33

1 (33%)

3 (9%)

0.20

2 (67%)

8 (24%)

0.12

1 (33%)

9 (27%)

0.82

Number and percentage (between brackets) of samples in which bacteria were detected. p-value (in bold and red when significant) as determined by Chi-Square analyses. Percentage calculated as percentage of the virus positive and virus negative value. *results from the OM2 cohort only

143

Coincident detection of bacteria and viruses was observed within the NPS

samples from children of the OM1 and OM2 cohorts (n=43) in addition to the

control cohort (n=17). Despite a range in the number and type of viruses detected

within the samples from each cohort, no significant correlations were observed

between virus and Spn, Hi or Mcat detections in these samples. (Table 4-8).


nasopharynx of children with and without RAOM/COME.

N Spn Hi Mcat

OM cohorts


4 39

1 (25%) 19 (49%)

0.37

2 (50%) 27 (69%)

0.43

1 (25%) 22 (56%)

0.23 ADV Pos ADV Neg p-value

5 38

3 (60%) 17 (45%)

0.52

3 (60%) 26 (68%)

0.7

4 (80%) 19 (50%)

0.21 hMPV Pos hMPV Neg p-value

3 40

3 (100%) 17 (43%)

0.05

2 (67%) 27 (68%)

0.98

2 (67%) 21 (53%)

0.64


16 27

8 (50%) 12 (44%)

0.72

11 (69%) 18 (67%)

0.89

9 (56%) 13 (48%)

0.78


8 35

4 (50%) 16 (46%)

0.32

6 (75%) 23 (66%)

0.61

5 (63%) 17 (49%)

0.57

OC43* Pos OC43 Neg p-value

3 17

1 (33%) 5 (29%)

0.89

3 (100%) 10 (59%)

0.17

3 (100%) 9 (53%)

0.13

EV* Pos EV Neg p-value

3 17

2 (67%) 4 (24%)

0.13

1 (33%) 12 (71%)

0.21

3 (100%) 9 (53%)

0.13

Control cohort


7 10

4 (57%) 3 (30%)

0.26

3 (43%) 7 (70%)

0.26

2 (29%) 6 (60%)

0.20


2 15

1 (50%) 6 (40%)

0.79

1 (50%) 9 (60%)

0.79

2 (100%) 6 (40%)

0.11


2 15

1 (50%) 6 (40%)

0.79

2 (100%) 8 (53%)

0.21

1 (50%) 7 (47%)

0.93 EV Pos EV Neg p-value

2 15

1 (50%) 6 (40%)

0.79

1 (50%) 9 (60%)

0.79

0 (0%) 8 (53%)

0.16

Number and percentage (between brackets) of samples in which bacteria were detected. p-value (in bold and red when significant) as determined by Chi-Square analyses. Percentage calculated as percentage of the virus positive and virus negative value. *results from the OM2 cohort only

144

Investigation of coincident virus and otopathogenic bacterial identifications within

the adenoid tissues from the OM2 and control cohorts (n=20 and n=16,

respectively) demonstrated significant positive correlations between the

presence of HboV and Mcat (r=0.56, p=0.01; Table 4-9) whilst EV presence was

negatively correlated with the detection of Spn (r=-0.48, p=0.03; Table 4-9) within

the OM2 cohort. Within the control group, identification of Spn was positively

correlated to detection of either ADV or HboV in the adenoids (Table 4-9).

145


adenoids of children with and without RAOM/COME.

N Spn Hi Mcat

OM2 cohort


13 7

8 (62%) 5 (71%)

0.66

11 (85%) 7 (100%)

0.27

9 (69%) 4 (57%)

0.59 HboV Pos HboV Neg p-value

13 7

9 (69%) 4 (57%)

0.589

12 (92%) 6 (86%)

0.639

11 (85%) 2 (29%)


14 6

7 (50%) 6 (100%)

0.03

13 (93%) 5 (83%)

0.52

11 (79%) 2 (33%)


7 13

6 (86%) 7 (54%)

0.15

7 (100%) 11 (85%)

0.27

5 (71%) 8 (62%)

0.66 HRV Pos HRV Neg p-value

17 3

11 (65%) 2 (67%)

0.95

16 (94%) 2 (67%)

0.14

12 (71%) 1 (33%)

0.21 PIV3 Pos PIV3 Neg p-value

11 9

7 (64%) 6 (67%)

0.89

9 (82%) 9 (100%)

0.18

8 (73%) 5 (56%)

0.42 OC43 Pos OC43 Neg p-value

4 16

3 (75%) 10 (63%)

0.64

4 (100%) 14 (88%)

0.46

4 (100%) 9 (56%)

0.10 WU Pos WU Neg p-value

11 9

8 (73%) 5 (56%)

0.42

10 (91%) 8 (89%)

0.88

8 (73%) 5 (56%)

0.42

Control cohort


7 9

7 (100%) 5 (56%)

0.04

7 (100%) 8 (89%)

0.36

4 (57%) 7 (78%)

0.38 HboV Pos HboV Neg p-value

11 5

10 (91%) 2 (40%)

0.03

10 (91%) 5 (100%)

0.49

9 (82%) 2 (40%)


10 6

6 (60%) 6 (100%)

0.07

10 (100%) 5 (83%)

0.18

6 (60%) 5 (83%)


4 12

3 (75%) 9 (75%)

1.00

4 (100%) 11 (92%)

0.55

2 (50%) 9 (75%)

0.35 HRV Pos HRV Neg p-value

11 5

9 (82%) 3 (60%)

0.35

10 (91%) 5 (100%)

0.49

7 (64%) 4 (80%)

0.51 PIV3 Pos PIV3 Neg p-value

8 8

6 (75%) 6 (75%)

1.00

8 (100%) 7 (88%)

0.30

5 (63%) 6 (75%)

0.59


7 9

5 (71%) 7 (78%)

0.77

7 (100%) 8 (89%)

0.36

4 (57%) 7 (78%)

0.38 WU Pos WU Neg p-value

5 11

4 (80%) 8 (73%)

0.76

5 (100%) 10 (91%)

0.49

3 (60%) 8 (73%)

0.61

Number and percentage (between brackets) of samples in which bacteria were detected. p-value (in bold and red when significant) as determined by Chi-Square analyses. Percentage calculated as percentage of the virus positive and virus negative value.

146

4.4 Discussion

Viral detections within the MEE of the present study occurred within 43% of MEE

samples. The presence of these viruses, within the middle ears of OM prone

children provides support to the possibility that OM prone children may have

established infectious reservoirs within the middle ear. Persistence of unresolving

infection may contribute to recurrent OM pathogenesis. The viral presence within

the middle ear of OM prone children contrasts to the sterile middle ear reported

in children and adults without OM (Westerberg et al., 2009).

The results of the current study are consistent with previous reports from Finland

(Pitkäranta et al., 1998), Turkey (Bulut et al., 2007), Japan (Monobe et al., 2003)

and the Netherlands (Stol et al., 2013). The frequency of detection reported in

this study are lower than a previous report from Western Australia, where viruses

were seen in 71% of MEE samples from young children with RAOM (Wiertsema

et al., 2011a). The higher rate of viral detection reported in the Western Australian

study may have resulted from examination of the viral prevalence in children of

younger age (<2 years) compared to the 4 year olds in the current study. The

most significant factor is likely to be the inclusion of children diagnosed with

COME in the present study who, unlike AOM patients, may not have active

infection at the time of surgery and collection of MEE.

Viral presence in the absence of any of the predominant bacterial otopathogens

occurred in 18% of MEE samples in this study, which is consistent with previous

reports using PCR for viral detection (Pitkäranta et al., 1998, Shinogami and

Ishibashi, 2004, Monobe et al., 2003, Stol et al., 2013). The high rate of viral

detections within the MEE support their potentially significant involvement in the

pathogenesis of OM (Chonmaitree et al., 2012). Indeed, nucleic acid from dead

microbes cannot last longer than 3 days in the middle ear fluid (Post et al., 1996).

Investigation of the interaction of viruses in OM pathogenesis using animal

models, has shown that IAV (without bacteria) caused AOM in 4% of studies

chinchillas (Giebink et al., 1980) and resulted in middle ear inflammation and

hearing loss in mice (Short et al., 2011). Furthermore, ADV can induce tympanic

membrane inflammation in the chinchilla middle ear (Bakaletz et al., 1993).

147

Animal studies have explored the rate and severity of OM pathogenesis

progressively from initial inoculation with the virus. There is increasing evidence

of similar effects in humans whereby identification of nasopharyngeal infection

with ADV increases the risk of the infected child developing AOM by up to 3-fold

(Binks et al., 2011). Evidence of an association between identification of ADV

infection and the risk of AOM development in young children has been recently

reported (Chonmaitree et al., 2015).

HRV was the predominant virus detected in all URT locations, including the

middle ear in children with RAOM/COME from both OM cohorts, despite their

recruitment over different periods (2008-2010 OM1 and 2015 OM2). A recent

study showed that detection of RHV nucleic acid in the respiratory tract is more

likely within 30 days of infection in infants (Loeffelholz et al., 2014). The

predominance of HRV is consistent with previous reports from Finland,

Netherland and Western Australia, where viral detection within the middle ear

was reported using PCR methodologies to examine the MEE (Pitkäranta et al.,

1998, Nokso-Koivisto et al., 2004, Ruohola et al., 2006, Wiertsema et al., 2011a,

Stol et al., 2013). In contrast, RSV was the most common virus identified (via

PCR) in the MEE of children with OM from Japan and Turkey (Monobe et al.,

2003, Bulut et al., 2007). Similar studies from the US also confirmed RSV as the

predominant virus within the middle ear, however these studies did not use PCR-

based detection methods (Heikkinen et al., 1999, Patel et al., 2007). Identification

of the predominant virus within the MEE of children experiencing OM clearly

varies with the region and recruitment criteria for participants in each study. Viral

incidence may vary at different times within a year or between years as evidenced

by a recent report from the US. The study by Makari et al., (2015) reported

significantly increased frequency of RSV detection in the URT of children with

OM in the peak season of RSV, compared to the shoulder seasons of RSV

(Makari et al., 2015). Therefore, the timing of recruitment maybe alter the study

population detection frequencies if it occurred during the low-activity RSV

seasons rather than the peak period (Chonmaitree, 2006). Each virus may

present differently over the year, for example, HRV infections occur year round

(Winther et al., 2006) which may increase the opportunity for detection and

reporting. Regardless of which virus is predominant, an effective vaccine against

148

that virus would benefit children with OM, who are positive for the virus,

unfortunately, for the two most commonly detected viruses, HRV and RSV,

effective vaccines are not currently approved or available for use (Glanville and

Johnston, 2015, Campbell et al., 2015).

In addition to variations in exposure to and/or infection with specific viruses over

the year, recruitment criteria may also impact the observed results. For example,

the child’s diagnosed presentation of OM whether AOM, RAOM and/or COME

may also influence the likelihood of the child experiencing active infection and

having virus/bacteria present, at the time of MEE collection. It is important to

recognise that all the previous reports cited have sampled MEE by clinical

perforation of the intact tympanic membrane. That is, the tympanic membrane

had not perforated as a result of OM pathogenesis. As discussed previously, the

more frequent use of PCR has improved the accuracy and reproducibility of viral

identification however increased scrutiny of the primers used to confirm these

identifications is beneficial to ensure similar identification characteristics.

Accurate and reproducible viral identifications and the sampling of multiple URT

locations within the same child have provided evidence of the nasopharynx being

the origin for viruses present in the middle ear (Winther et al., 2007, Alper et al.,

2009, Chonmaitree et al., 2008a). Specifically within Australia, this study of

children from Eastern Australia, provides results consistent with a prior report

from Western Australia, in which children with RAOM had more frequent

identification of viral detections in the nasopharynx compared to healthy control

children (Wiertsema et al., 2011a). In the current study, the presence of HRV,

IBV and OC43 within the nasopharynx was positively associated with the

presence of the same viruses in the middle ear of the same child with

RAOM/COME. Similar results were observed for other viruses, including ADV,

RSV, hMPV and WU, identified in the middle ear and nasopharynx of children

with RAOM/COME more often than it was observed in the NPS of children without

RAOM/COME. Interestingly, co-occurrence of a number of specific viruses in the

nasopharynx and adenoids including HRV, hMPV, PIV3 and ADV was observed

more frequently in children with RAOM/COME than those without, and this may

indicate that the presence of multiple viruses within the URT could increase the

149

risk of OM. Collectively, the results confirm that the presence of viruses within the

URT increases the risk of detection within the middle ear and their association

with OM pathogenesis.

The frequency of viral detection varied by location within the URT. Specifically,

all adenoid tissues from children with RAOM/COME and those of the control

group, undergoing treatment for adenoidal hypertrophy (AH) and/or obstructive

sleep apnoea (OSA) had at least one virus present. This finding is consistent with

previous studies whereby 90-100% adenoidal samples from children with AH,

with RAOM, or with OME had viruses detected (Sato et al., 2009, Herberhold et

al., 2009, Szalmás et al., 2013). HboV detection in the adenoids of children

occurred at higher frequencies in children with OME compared to children with

RAOM or OSA only (Szalmás et al., 2013). The prevalence of HboV

identifications did not differ between children with or without RAOM/COME,

however the limitation of the current study’s recruitment, whereby RAOM and

COME were not specifically differentiated in the older (4 year old) children may

have affected these analyses. Although the current study identified HRV as the

most frequently detected virus within the adenoids, there was no significant

difference between children with or without RAOM/COME. This contrasts to a

previous report whereby HRV was more frequently detected in children who had

MEE present at the time of adenoidectomy (Rihkanen et al., 2004). Differences

in study design, including the age of the children (1 year older in current study),

viral detection methods used (PCR in current study) and the cohort size may

contribute to the variation between these results. Detection of WU polyomavirus

was prevalent in the adenoids, regardless of the patient’s OM history and the high

frequency of detection is consistent with a previous report from children

undergoing adenoidectomy to treat AH (Comar et al., 2011).

The frequent identification of a range of viruses including ADV, EV, OC43, HRV

and WU, and the strong correlation observed for their co-detection in the

nasopharynx and adenoids of the same children, is supportive of these regions

acting as potential reservoirs for viral infection of the URT. Nasopharyngeal

prevalence of viruses reportedly decreased following adenoidectomy (Biill Primo

150

et al., 2014), which further evidences the involvement of both the nasopharynx

and adenoids in URT infection.

Regardless of the location of the viral infection, this study identified that

concurrent identification of viruses and bacteria occurred in 24% of MEE

samples, a rate lower than was observed in a previous report from children with

AOM (66%) in which PCR methodology was used (Ruohola et al., 2006). The use

of PCR to improve the accuracy and reproducibility of viral detection in MEE

samples in this study was similar to the frequency of co-incident identification

reported in a previous report (15%), which used high quality viral culture methods.

That report, similar to the study from Western Australia (Wiertsema et al., 2011a)

reported the incidence within children experiencing AOM, rather than

RAOM/COME (Chonmaitree, 2000).

The role of co-occurrence of specific microbes in potentially increasing the risk of

OM pathogenesis was demonstrated in the present study by the co-detection and

correlation of ADV and Hi within RAOM/COME patients in the current study.

Further exploration of co-infection and its impact on OM pathogenesis has been

provided through animal models, with elegant research using the chinchilla often

considered the most representative of the human condition. Using the chinchilla

model, Suzuki and Bakaletz (1994) demonstrated that intranasal inoculation with

both ADV type 1 and NTHi resulted in development of more severe OM than was

observed using either agent as a single pathogen (Suzuki and Bakaletz, 1994,

Miyamoto and Bakaletz, 1997). In addition, intranasal inoculation of chinchillas

using ADV may increase the transduction of NTHi from the nasopharynx to the

middle ear and induce NTHi OM (Miyamoto and Bakaletz, 1997). Furthermore,

using mouse models, Shorts et al. (2011) has demonstrated that infection with

influenza virus enable development and persistence of pneumococcal OM via

neutrophil extracellular trap induction (Short et al., 2011), which is important in

the persistence of effusion and bacteria in children with RAOM (Thornton et al.,

2013). Taken together, it appears that viral-bacterial co-infection may play

significant roles in development and persistence of OM.

This is the first study using multiplex real-time PCR to identify a wide range of

viruses in different locations of the URT from children with and without

151

RAOM/COME. HRV was still the most common virus detected in the middle ear

and nasopharynx of urban children with RAOM/COME in Eastern Australia and

all HRV detected in the middle ear were also detected in the nasopharynx of the

same child. Importantly, children with RAOM/COME had higher frequency of

viruses detected in the nasopharynx, compared with those without

RAOM/COME. Significantly, viruses were present within the adenoids and NPS

of the URT for all children in the study, whether prone to OM (ie RAOM/COME

patients) or control patients (AH/OSA). Clearly, viral presence alone does not

stimulate OM pathogenesis, but evidence of co-infection with predominant

bacterial otopathogens did increase the risk of OM pathogenesis in

RAOM/COME children compared to controls.

The clear association between viral and bacterial pathogens causal for OM

pathogenesis reinforces the importance of consideration of the multifactorial

causality of OM pathogenesis. Further investigation of these relationships

between the most common viruses and bacteria within the URT, using larger

cohort sizes than was possible in the current study, will provide evidence to guide

vaccine development, to optimise the opportunity to prevent rather than treat OM

in vulnerable populations where other factors, such as living standards or access

to health care may not be as rapidly or readily improved.

152

Supplemental Table 4-1: Comparison of viral detection amongst different locations of the upper respiratory tract of children within the OM1, OM2 and control cohorts.

OM1 cohort OM2 cohort Control cohort

AD NPS MEE p-value

AD vs NPS

p-value

AD vs MEE

p-value

NPS vs MEE

AD NPS NS MEE p-value AD vs NPS

p-value

AD vs NS

p-value

AD vs MEE

p-value

NPS vs NS

p-value

NPS vs MEE

p-value NS vs MEE

AD NPS NS p-value AD vs NPS

p-value

AD vs NS

p-value

NPS vs NS

OM1

(n = 19)

OM1

(n = 23)

OM1

(n = 45)

OM2

(n = 20)

OM 2

(n = 20)

OM2

(n = 20)

OM2

(n = 36)*

Control

(n = 16)

Control

(n = 17)

Control

(n = 16)

Children positive for at least one studied virus

19 (100%)

16 (70%)

20 (44%)

0.008 0.000 0.05 20 (100%)

13 (65%)

8 (40%)

15 (42%)

0.004 0.000 0.000 0.11 0.09 0.90 16 (100%)

8 (47%)

6 (38%)

0.001 0.000 0.58

Adenovirus 6 (32%) 2 (9%) 3 (7%) 0.06 0.009 0.76 13 (65%)

3 (15%)

0 (0%) 0 (0%) 0.001 0.000 0.000 0.07 0.02 na 7 (44%) 0 (0%) 0 (0%) 0.002 0.003 na

Bocavirus na na na na na Na 13 (65%)

0 (0%) 0 (0%) 3 (8%) 0.000 0.000 0.000 na 0.18 0.18 11 (69%)

0 (0%) 0 (0%) 0.000 0.000 na

Coronavirus

229E

HKU-1

NL63

OC43

na

na

na

na

na

na

na

na

na

na

na

na

na

na

na

na

na

na

na

na

na

na

na

na

0 (0%)

0 (0%)

2 (10%)

4 (20%)

0 (0%)

0 (0%)

0 (0%)

3 (15%)

0 (0%)

0 (0%)

0 (0%)

1 (5%)

0 (0%)

0 (0%)

0 (0%)

1 (3%)

na

na

0.15

0.68

na

na

0.15

0.151

na

na

0.05

0.03

na

na

na

0.29

na

na

na

0.09

na

na

na

0.67

0 (0%)

0 (0%)

1 (6%)

7 (44%)

0 (0%)

0 (0%)

0 (0%)

2 (12%)

0 (0%)

0 (0%)

0 (0%)

2 (13%)

na

na

0.30

0.04

na

na

0.31

0.05

na

na

na

0.95

Enterovirus na na na na na na 14 (70%)

3 (15%)

0 (0%) 0 (0%) 0.000 0.000 0.000 0.07 0.02 na 10 (63%)

2 (12%)

0 (0%) 0.002 0.000 0.16

Human metapneumovirus 0 (0%) 3 (13%)

2 (4%) 0.10 0.35 0.20 7 (35%) 0 (0%) 0 (0%) 1 (3%) 0.004 0.004 0.001 na 0.45 0.45 4 (25%) 0 (0%) 0 (0%) 0.03 0.03 na

Influenza virus

IAV

IBV

0 (0%)

0 (0%)

1 (4%)

0 (0%)

0 (0%)

0 (0%)

0.36

na

na

na

0.16

na

1 (5%)

0 (0%)

0 (0%)

1 (5%)

0 (0%)

0 (0%)

0 (0%)

1 (3%)

0.31

0.31

0.31

na

0.18

0.45

na

0.31

na

0.67

na

0.45

0 (0%)

0 (0%)

0 (0%)

0 (0%)

0 (0%)

0 (0%)

na

na

na

na

na

na

Respiratory syncytial virus 0 (0%) 1 (4%) 3 (7%) 0.36 0.25 0.70 1 (5%) 1 (5%) 0 (0%) 3 (8%) 1.00 0.31 0.64 0.31 0.64 0 (0%) 0 (0%) 0 (0%)

Rhinovirus 10 (53%)

7 (30%)

10 (22%)

0.15 0.02 0.46 17 (85%)

9 (45%)

6 (30%)

6 (17%)

0.008 0.000 0.000 0.33 0.02 0.24 11 (69%)

7 (41%)

4 (25%)

0.11 0.01 0.33

Parainfluenza virus

PIV1

PIV2

PIV3

0 (0%)

2 (11%)

2 (11%)

0 (0%)

4 (17%)

0 (0%)

0 (0%)

0 (0%)

0 (0%)

na

0.53

0.11

na

0.03

0.03

na

0.004

na

3 (15%)

2 (10%)

11 (55%)

0 (0%)

0 (0%)

1 (5%)

0 (0%)

1 (5%)

0 (0%)

0 (0%)

0 (0%)

0 (0%)

0.07

0.15

0.001

0.07

0.55

0.000

0.02

0.05

0.000

na

0.31

0.31

na

na

0.18

na

0.18

na

1 (6%)

5 (31%)

8 (50%)

0 (0%)

1 (6%)

0 (0%)

0 (0%)

0 (0%)

0 (0%)

0.30

0.06

0.001

0.31

0.02

0.001

na

0.33

na

Polyomavirus

WU

KI

11 (58%)

na

7 (30%)

na

4 (9%)

na

0.07

na

0.000

na

0.02

na

11 (55%)

3 (15%)

1 (5%)

1 (5%)

0 (0%)

0 (0%)

1 (3%)

0 (0%)

0.001

0.29

0.000

0.07

0.000

0.02

0.31

0.31

0.67

0.18

0.45

na

5 (31%)

2 (13%)

0 (0%)

2 (12%)

0 (0%)

0 (0%)

0.01

0.95

0.02

0.11

na

0.16

Single virus 12 (63%)

9 (39%)

19 (42%)

0.12 0.13 0.81 0 (0%) 5 (25%)

8 (40%)

14 (39%)

0.02 0.001 0.001 0.31 0.29 0.94 1 (6%) 4 (24%)

6 (38%)

0.17 0.03 0.38

Multiple viruses 7 (37%) 7 (30%)

1 (2%) 0.66 0.000 0.001 20 (100%)

8 (40%)

0 (0%) 1 (3%) 0.000 0.000 0.000 0.004 0.001 0.45 15 (94%)

4 (24%)

0 (0%) 0.000 0.000 0.04

Pearson Chi-square was used to compare frequencies of detection in each of the specimens; na: not available

153

Chapter 5:

Immune responses in urban children with

and without RAOM/COME in South East

Queensland

154

5.1 Introduction

Pathogen specific immune deficiencies, both humoral and cellular, may exist in

children with OM making them more susceptible to both colonisation and

subsequent infections in the middle ear space. Within the literature, there are

conflicting data on the immune responses of these children who are prone to

recurrent and/or chronic episodes of OM, and in particular antibody levels to

specific antigens of the two main bacterial otopathogens, Spn and Hi. Some

groups have indicated that children who are classified as otitis prone have

significantly lower levels of serum IgG and IgA to pneumococcal serotypes and

proteins (Freijd et al., 1984, Veenhoven et al., 2004, Prellner et al., 1989, Dhooge

et al., 2002, Kaur et al., 2011a) as well as to proteins from Hi (Pichichero et al.,

2010, Kaur et al., 2011b) when compared non-otitis prone children or healthy

controls. In contrast, several investigators have demonstrated the same or higher

levels of serum IgG, IgA, and its subclasses to pneumococcal polysaccharides

(Berman et al., 1992, Drake-Lee et al., 2003, Misbah et al., 1997, Sørensen and

Nielsen, 1988, Menon et al., 2012, Corscadden et al., 2013) and proteins

(Wiertsema et al., 2012) or Hi proteins (Wiertsema et al., 2012) in children with

OM, compared to healthy control children. The conflicting data observed for

antibody titres in these studies are likely due to both population and study design

differences. Study design differences include challenges with appropriate

selection of “healthy” control groups. A common surgical control group recruited

for studies in OM include children undergoing surgery for adenoidal or

adenotonsillar hypertrophy. While these children are not “healthy” they represent

the only population from which adenoidal tissue can be removed and analysed.

This study examines local and systemic pathogen specific antibody to

pneumococcal (PspA1, PspA2, Ply and CbpA) and Hi (P26, P4, P6 and PD)

protein vaccine candidates in children with and without RAOM/COME.

Cellular immune responses, including cytokine production and

frequency/competency of cell types, may play a role in a child’s susceptibility to

OM (Smirnova et al., 2002, Skotnicka and Hassmann, 2000, Zhao et al., 2009).

An example of this is that high levels of TNF-α, IL-8, IL-10 and the presence of

Th2 cytokines IL-4, IL-5 and IL-13 in MEE appear to correlate with the persistence

155

of effusions in OME and are thought to relate to the chronicity of infection (Yellon

et al., 1991, Kubba et al., 2000, Smirnova et al., 2005, Zhao et al., 2009).

The balance of cytokine production by specific immune cells, including T helper

cells (Th: Th1, Th2 and Th17) and regulatory T (Treg) cells, facilitate the host

response to bacterial and viral infection (Couper et al., 2008, Mangodt et al.,

2015). Furthermore, Treg cells play a vital role in immune regulation and are

thought to be linked to have an immunosuppressive function (Sakaguchi et al.,

2009). Decreased numbers, or defective function of Treg cells are implicated in

the development of autoimmunity or asthma (Suzuki et al., 2011, Ryanna et al.,

2009). Conversely, increased numbers of Treg cells are believed to contribute to

chronic and persistent infectious diseases, such as chronic hepatitis C virus

infection and chronic herpes simplex virus infection (Mills, 2004, Keynan et al.,

2008). Interestingly, increased frequencies of Treg cells (CD4+CD25+FoxP3+ T

cells) have been observed in patients with OM compared to healthy controls and

are thought to be associated with the chronicity of OME (Zhao et al., 2009). In

this study, cellular immune responses were assessed by examination of

cytokines IL-1β, IL-5, IL-6, IL-8, IL10, IL-12p70, IL-13, IL-17, IFN-γ and TNFα in

the MEE and serum of children with RAOM/COME and in serum of children

without RAOM/COME, in addition to analysis of Treg populations in peripheral

mononuclear cells.

Immunological data from this study will provide evidence on immune responses

in a defined population of children with RAOM/COME, to help inform vaccine

development and targets for OM prevention.

5.2 Methods

Patient recruitment, sample collection and processing, serum collection, antibody

assays, cytokine assays, PBMC isolation and Treg cell identification were

performed as described in Chapter 2, sections 1, 2, 10, 11, 12, 13, and 14,

respectively.

Serum antibody titres were reported in arbitrary units per millilitre (ml) of serum

(AU/ml). MEE and saliva antibody data were standardised against total protein,

quantified using the bicinchoninic acid (BCA) assay from each sample, to correct

156

for sample collection effects. Antibody was reported in AU per milligram (mg) of

total protein (AU/mg) per ml of sample. Similarly, MEE cytokine data were

normalised with total protein in each sample and reported in picogram per

milligram of total protein (pg/mg) per ml of sample.

Host and environmental risk factors were compared between children from the

OM control cohort using Student’s t tests for continuous variables (age and

number of episodes of OM) and Pearson Chi-Square analyses (p-value

asymptotic significant 2-sided) for categorical variables (gender, attendance at

day-care, antibiotic use, siblings, breastfeeding, Pacifier use, cigarette exposure

and vaccination status). Antibody levels were reported as antigen-specific

geometric mean concentrations (GMCs) with 95% confidence intervals (CI).

Similarly, cytokine levels were reported as GMCs with 95% CI. Antibody and

cytokine levels, as well as proportion of T cell population, were compared

between groups using Student’s t tests. Linear regression analyses were used to

examine correlation in antigen-specific antibody levels in serum, MEE and saliva.

Similarly, linear regression analyses were performed to examine correlation in

levels of different cytokines present in MEE. A p-value <0.05 was considered

statistically significant. The IBM SPSS Statistics 23 for Windows software

package (IBM, New York, USA) was used for all statistical analyses and data

were plotted using GraphPad Prism 5.0 (Graph Software Inc., California, USA).

157

5.3 Results


Twenty children with RAOM/COME (OM2 cohort) and 17 children without

RAOM/COME (control cohort) were recruited prior to time of surgery for VTI for

RAOM/COME or adenoidectomy for AH/OSA, respectively (described in Chapter

3, section 3.1) (Table 5-1). Thirty-four MEE samples were analysed (6 samples

had insufficient volume) from children with RAOM/COME. Serum samples were

analysed for all 37 children. Only 11 salivary samples (5 from OM2 and 6 from

controls) were analysed, due to insufficient volumes in the samples collected from

the remainder of each cohort.

No significant differences in host factors, such as age, gender, allergy and

asthma, as well as environment factors, as such, attending day-care, having

siblings and being exposed to cigarette smoke between OM and control children

(Table 5-1).

158

Table 5-1: Characteristic demographic data for children in each cohort of this

study.

OM2 cohort

Control cohort

p-value

Number 20 17

Mean age in year (range) 4.1 (2–8) 4.4 (3–8) 0.60

Male (%) 11 (55%) 12 (71%) 0.51


7.1 (4-11) 0.9 (0-2) <0.001

Attending day-care 18/19 (95%) 13/16 (81%) 0.21

Antibiotic use in the last month 11/19 (58%) 7/16 (44%) 0.40

Having siblings 17/19 (89%) 15/16 (94%) 0.65

A history of breastfeeding 16/19 (84%) 14/16 (88%) 0.78

A history of pacifier use 11/19 (58%) 7/16 (44%) 0.40

Allergy to any substance a 4b/19 (21%) 4c/16 (25%) 0.78

Asthma 2/19 (11%) 4/16 (25%) 0.26

Exposed to cigarette smoke 6/19 (32%) 4/16 (25%) 0.67

Fully vaccinated with PCV7 9 (45%) 9 (53%) 0.63

Fully vaccinated with PCV13 11 (55%) 8 (47%) 0.63

a Cow’s milk protein, egg, wheat, shellfish, house dust mite, pollens/trees, moulds, grasses, bee/insect venom, peanut, other nuts, cat/dog dander, others. b Allergic to penicillin (n=2), strawberries and kiwi fruit (n=1) and to cow’s milk protein (n=1).

c Allergic to penicillin (n=1), wheat (n=1), gluten (n=1) and cow’s milk protein (n=1).


Pearson Chi-Square analyses were used categorical variables (Male, attending day-care, antibiotics use, siblings, breastfeeding, pacifier use, allergy, asthma, cigarette smoke expose and vaccination)

159

5.3.2 Serum immune responses in children with and without

RAOM/COME

IgG and IgA antibodies to PspA1, PspA2, CbpA, Ply (Spn), P26, P4, P6 and PD

(Hi) were measured in serum samples from 20 children with RAOM/COME and

17 children without RAOM/COME.

5.3.2.1 Pneumococcal protein specific IgG and IgA antibody levels in

serum of children with and without RAOM/COME.

No differences in serum IgG or IgA titres to any of the 4 pneumococcal proteins

were seen between children with and without RAOM/COME (Figure 5-1a and b).

160

Serum IgG to Spn proteinsIg

G (

AU

/ml)

Psp

A1

Psp

A2

Ply

CbpA

1

10

100

1000

10000

100000

OM

Control

a

Serum IgA to Spn proteins

IgA

(A

U/m

l)

Psp

A1

Psp

A2

Ply

CbpA

1

10

100

1000

OM

Control

b

Figure 5-1: Geometric mean titres of IgG and IgA to Spn proteins in serum of

children with and without RAOM/COME. OM group (n = 20) and control group (n

= 17). Data were analysed using Student’s t tests on log-transformed data and

values represent GMC with 95% CI.

161

5.3.2.2 Hi protein specific IgG and IgA antibody levels in serum of


No differences in levels of anti-Hi protein IgG were observed in serum from

children with or without OM (Figure 5-2a). Anti-P4 IgA was significantly higher in

the serum of children with RAOM/COME compared to those of the control cohort

(p=0.03; Figure 5-2b). No significant differences in serum anti-P26, -P6 and -PD

IgA were observed between the groups (Figure 5-2b).

162

Serum IgG to Hi proteinsIg

G (

AU

/ml)

P26 P

4P6

PD

1

10

100

1000

10000

100000

OM

Control

a

Serum IgA to Hi proteins

IgA

(A

U/m

l)

P26 P

4P6

PD

1

10

100

1000

10000

OM

Control

p=0.034

b

Figure 5-2: Geometric mean titres of IgG and IgA to Hi proteins in serum of

children with and without RAOM/COME. OM group (n = 20) and control group (n

= 17). Data were analysed using Student’s t tests on log-transformed data and

values represent GMC with 95% CI.

163

5.3.3 Serum antibody levels according to nasopharyngeal

colonisation in children with and without RAOM/COME

The effects of nasopharyngeal colonisation with Spn or Hi on antibody titres in

children with RAOM/COME (6 and 13 respectively) versus control children (7 and

10 respectively) were examined. All nasopharyngeal samples that were culture

positive were also PCR positive, thus the results reported are based on PCR.

5.3.3.1 Serum antibody titres to pneumococcal proteins in children with

and without Spn colonisation.

No differences in serum anti-pneumococcal protein IgG and IgA titres were

observed between children with or without Spn colonisation for either the OM or

control cohort (Figure 5-3).

164

Serum IgG - OM cohort

IgG

(A

U/m

l)

Psp

A1

Psp

A2

Ply

CbpA

1

10

100

1000

10000

100000

aSerum IgG - Control cohort

IgG

(A

U/m

l)

Psp

A1

Psp

A2

Ply

CbpA

1

10

100

1000

10000

100000

b

Serum IgA - OM cohort

IgA

(A

U/m

l)

Psp

A1

Psp

A2

Ply

CbpA

1

10

100

1000

10000

cSerum IgA - Control cohort

IgA

(A

U/m

l)

Psp

A1

Psp

A2

Ply

CbpA

1

10

100

1000

10000

d

Figure 5-3: Geometric mean titres of IgG and IgA to Spn proteins in serum from children with and without Spn nasopharyngeal

colonisation within the OM and control cohorts. Data were analysed using Student’s t tests on log-transformed data and values

represent GMC with 95% CI.

Colonisation with Spn

Not colonisation with Spn

165

5.3.3.2 Serum antibody titres to Hi proteins in children with and without Hi

colonisation

No differences were observed for serum IgG to any of the Hi proteins between children

with and without Hi colonisation in the OM group (Figure 5-4a). However, anti-P4 IgA

was significantly higher in the serum of children with Hi colonisation than those without

in the OM cohort (p=0.02; Figure 5-4c). No differences were observed for serum anti-

P26, -P6 and –PD IgA for children on the basis of colonisation in the OM cohort

(Figure 5-4c). No differences in IgA or IgG titres were observed to any of the Hi

proteins in the control cohort based on colonisation status (Figure 5-4b and d).

166

Serum IgG - OM cohort

IgG

(A

U/m

l)

P26 P

4P6

PD

1

10

100

1000

10000

100000

a Serum IgG - control cohort

IgG

(A

U/m

l)

P26 P

4P6

PD

1

10

100

1000

10000

100000

b

Serum IgA - OM cohort

IgA

(A

U/m

l)

P26 P

4P6

PD

1

10

100

1000

10000 p=0.02

c Serum IgA - control cohort

IgA

(A

U/m

l)

P26 P

4P6

PD

1

10

100

1000

10000

100000

d

Figure 5-4: Geometric mean titres of IgG and IgA to Hi proteins in serum from children with and without Hi colonisation in nasopharynx

within the OM and control cohorts. Data were analysed using Student’s t tests on log-transformed data and values represent GMC

with 95% CI.

Colonisation with Hi

Not colonisation with Hi

167

5.3.4 Serum antibody levels based on Spn and Hi in the middle ear of

children with RAOM/COME

Spn and Hi specific serum antibody titres were compared between children with MEE

positive and negative for Spn (n=4) and Hi (n=6). The presence and absence of Spn

and Hi in MEE were reported based on PCR. If either the left or right ear of the same

children was positive for these bacteria the child was considered as having a positive

MEE.

Serum anti- PspA1, -PspA2 and –CbpA IgA were higher in children with Spn positive

MEE (p=0.004, p=0.03 and p=0.01, respectively; Figure 5-5b). No differences in IgG

to any other proteins were observed, based on Spn detection in the MEE (Figure 5-

5a). No differences in serum IgG or IgA were seen for any of the Hi proteins, based

on Hi detection in the MEE (Figure 5-5c and d).

168

Serum IgG to Spn proteins

IgG

(A

U/m

g)

Psp

A1

Psp

A2

Ply

CbpA

1

10

100

1000

10000

100000

1000000

a Serum IgA to Spn proteins

IgA

(A

U/m

g)

Psp

A1

Psp

A2

Ply

CbpA

1

10

100

1000

10000

100000P=0.004

P=0.03 P=0.01

b

Serum IgG to Hi proteins

IgG

(A

U/m

g)

P26 P

4P6

PD

1

10

100

1000

10000

100000

cSerum IgA to Hi proteins

IgA

(A

U/m

g)

P26 P

4P6

PD

1

10

100

1000

10000

d

Figure 5-5: Geometric mean titres for serum IgG and IgA to Spn and Hi proteins between children with MEE positive or negative for

bacteria in the OM cohort. Data were analysed using Student’s t tests on log-transformed data and values represent GMC with 95%

CI.

MEE positive for Spn

MEE negative for Spn

MEE positive for Hi

MEE negative for Hi

169

5.3.5 Mucosal antibodies against pneumococcal and Hi proteins in


Antibodies were measured in the MEE (n=34) and saliva (n=5) of children undergoing

surgery for RAOM/COME and in the saliva of control (n=6) children. These were

assessed according to PCR detection of the target bacterial otopathogens, Spn (n=4)

and Hi (n=10), in the MEE (Chapter 3).

5.3.5.1 Pneumococcal specific IgA and IgG in the MEE of children with

RAOM/COME

IgA and IgG to PspA1, PspA2 and Ply and IgG to CbpA were assessed. No differences

in antibody titres were seen between those MEE which were positive or negative for

Spn (Figure 5-6a and b).

Frequency of anti-PspA1 IgA detection was significantly higher in MEE negative for

Spn than in MEE positive for Spn (p=0.005; Table 5-2). No significant differences in

frequencies of detection of other antibodies between MEE positive and negative for

Spn (Table 5-2).

Table 5-2: Frequencies of MEE samples with anti-pneumococcal protein IgA and IgG

titres above the limit of detection (LOD) from children in the OM2 cohort.

Antibody in MEE Total MEE

(n = 34)


(n = 30)

MEE positive for Spn

(n = 4)

p-value a

IgG PspA1 27 (79%) 24 (80%) 3 (75%) 0.94

PspA2 34 (100%) 30 (100%) 4 (100%) na

Ply 34 (100%) 30 (100%) 4 (100%) na

CbpA 34 (100%) 30 (100%) 4 (100%) na

IgA PspA1 33 (97%) 30 (100%) 3 (75%) 0.005

PspA2 20 (59%) 18 (60%) 2 (50%) 0.70

Ply 24 (71%) 20 (67%) 4 (100%) 0.17

a p-value in comparison of frequencies of antibody detection between MEE positive and negative for Spn using Pearson Chi-square analysis.

170

MEE IgG to Spn proteins

IgG

(A

U/m

g)

Psp

A1

Psp

A2

Ply

CbpA

1

10

100

1000

10000

100000MEE positive for Spn


a

MEE IgA to Spn proteins

IgA

(A

U/m

g)

Psp

A1

Psp

A2

Ply

0.1

1

10

100

1000

10000MEE positive for Spn


b

Figure 5-6: Geometric mean IgG and IgA titres to PspA1, PspA2, Ply and CbpA in

MEE from children with RAOM/COME. Data were analysed using Student’s t tests on

log-transformed data and values represent GMC with 95% CI.

171

5.3.5.2 IgG and IgA to Hi protein antigens in the MEE of children with

RAOM/COME

Frequency of anti-P26 IgG detection was significantly higher in MEE negative for Hi

than in MEE positive for Hi (p=0.03; Table 5-3). No difference in frequencies of

detection of other antibodies against Hi proteins between MEE positive and negative

for Hi (Table 5-3).

Table 5-3: The number of MEE samples having anti-Hi protein IgG and IgA titres

higher than limit of detection (LOD) from children of OM2 cohort.

Antibody in MEE

Total MEE

(n = 34)

MEE negative for Hi

(n = 24)

MEE positive for Hi

(n =10)

p-value a

IgG P26 22 (65%) 18 (75%) 4 (40%) 0.03

P4 34 (100%) 24 (100%) 10 (100%) na

P6 34 (100%) 24 (100%) 10 (100%) na

PD 29 (85%) 22 (92) 7 (70%) 0.10

IgA P26 26 (76%) 20 (83%) 6 (60%) 0.14

P4 20 (59%) 16 (67%) 4 (40%) 0.15

P6 27 (79%) 20 (83%) 7 (70%) 0.38

PD 29 (85%) 22 (92) 7 (70%) 0.10

a p-value in comparison of frequencies of antibody detection between MEE positive and negative for Hi using Pearson Chi-square analysis.

No significant differences were observed for IgG titres between those MEE which were

positive or negative for Hi detection (Figure 5-7a). Anti-P6 IgA was higher in those

MEE samples that were negative for Hi (p=0.01), however no other significant

differences were seen (Figure 5-6b).

172

MEE IgG to Hi proteins

IgG

(A

U/m

g)

P26 P

4P6

PD

1

10

100

1000

10000MEE positive for Hi

MEE negative for Hi

a

MEE IgA to Hi proteins

IgA

(A

U/m

g)

P26 P

4P6

PD

1

10

100

1000MEE positive for HiMEE negative for Hi

p=0.013

b

Figure 5-7: Geometric mean IgG and IgA titres against P26, P4, P6 and PD in MEE

from children with RAOM/COME. Data were analysed using Student’s t tests on log-

transformed data and values represent GMC with 95% CI.

173

5.3.5.3 Antibody titres in saliva of children with and without RAOM/COME

IgA and IgG to Spn and Hi proteins were measured in 5 saliva samples from children

with RAOM/COME and 6 samples from children without RAOM/COME. IgA titres for

7 proteins (PspA1, PspA2, Ply, P26, P4, P6 and PD) were higher than LOD in all 11

saliva samples from both groups of children (Table 5-4). In contrast, IgG titres for 8

proteins (PspA1, PspA2, Ply, CbpA, P26, P4, P6 and PD) higher than LOD were

detected in 33% to 100% of saliva samples (Table 5-4). No significant differences in

frequencies of detection of any tested antibody in saliva between the OM and control

cohorts (Figure 5-4).

Table 5-4: The number of saliva samples having anti Spn and Hi protein IgG and IgA

titres higher than limit of detection (LOD) from both OM and control cohorts.

Antibody in saliva OM cohort

(n = 5)

Control cohort

(n = 6)

p-value a

IgG PspA1 3 (60%) 4 (67%) 0.82

PspA2 5 (100%) 5 (83%) 0.34

Ply 5 (100%) 5 (83%) 0.34

CbpA 5 (100%) 4 (67%) 0.15

P26 4 (80%) 2 (33%) 0.12

P4 3 (60%) 5 (83%) 0.39

P6 5 (100%) 4 (67%) 0.15

PD 4 (80%) 4 (67%) 0.62

IgA PspA1 5 (100%) 6 (100%) na

PspA2 5 (100%) 6 (100%) na

Ply 5 (100%) 6 (100%) na

P26 5 (100%) 6 (100%) na

P4 5 (100%) 6 (100%) na

P6 5 (100%) 6 (100%) na

PD 5 (100%) 6 (100%) na

a p-value in comparison of frequencies of antibody detection in saliva of children between the OM and control cohorts.

Salivary anti-P26 IgG titres were significantly higher in children with RAOM/COME

compared to the control cohort (p=0.03; Figure 5-8a). IgG to the remaining Hi proteins

were not different between the groups (Figure 5-8a). No differences in the IgA titres

to any of the Hi proteins assessed were seen between children with OM and their

control counterparts (Figure 5-8b).

174

Saliva IgA to Spn and Hi proteins

IgA

(A

U/m

g)

Psp

A1

Psp

A2

Ply

P26 P

4P6

PD

1

10

100

1000

10000

OMControl

b

Saliva IgG to Spn and Hi proteinsIg

G (

AU

/mg

)

Psp

A1

Psp

A2

Ply

CbpA

P26 P

4P6

PD

0.1

1

10

100

OMControl

p=0.033

a

Figure 5-8: Salivary IgG and IgA against Spn and Hi proteins in children with and

without RAOM/COME. Comparisons were performed using Student’s t tests on log-

transformed data. Values represent GMC with 95% CI.

175

5.3.5.4 Correlation in levels of systemic (serum) and local (MEE and saliva)

antibodies in children with and without RAOM/COME

To determine if local antibody production differed between the left and right ears of the

same child, correlations were performed on 17 pairs of MEE. No correlations for

antigen specific IgA titres in the left and right ear of the same child were observed

(Table 5-5). In contrast, strong correlations in anti-P26, -P4, -P6 and –PspA2 IgG titres

were observed between MEE from the left and right ears of the same children (Table

5-5).

As IgG titres were observed to correlate strongly between the ears of the same child,

an average was calculated and used to perform correlations with serum and salivary

antibodies. IgG against P26, P6 and PspA2 were observed to correlate strongly

between the MEE and serum of children with RAOM/COME (Table 5-5). When

correlations between MEE and salivary antibodies were performed, strong correlation

for IgG to P4 and P6 were also observed (Table 5-5).

Antibody titres from the serum and saliva of 11 children (with and without

RAOM/COME) were correlated to assess the site of antibody production. Strong

correlations were observed for IgA against P4, P26, PspA1 and 2 and Ply were

observed (Table 5-5). Correlations for IgA against CbpA could not be performed as

this protein binds secretory component and the assay is thus not specific in mucosal

samples. When correlations for IgG between the sites were performed, antibody

against both PspA1 and PspA2 were seen to be significantly correlated (Table 5-5).

176

Table 5-5: Correlation in levels of specific antibodies in MEE, saliva and serum of children from the OM and control cohorts.

Antibody R (left vs right ear)

(17 pairs)a

p-value

R (MEE vs serum)

(17 pairs)a

p-value

R (MEE vs saliva)

(5 pairs)a

p-value

R (Saliva vs serum)

(11 pairs)b

p-value

IgA P26 0.05 0.86 na na na na 0.91 0.000

P4 0.06 0.82 na na na na 0.89 0.000

P6 0.25 0.34 na na na na 0.3 0.35

PD 0.04 0.89 na na na na 0.22 0.5

PspA1 0.05 0.85 na na na na 0.95 0.000

PspA2 0.11 0.67 na na na na 0.89 0.000

Ply 0.22 0.4 na na na na 0.72 0.008

IgG P26 0.86 0.000 0.79 0.000 0.33 0.58 0.29 0.36

P4 0.54 0.02 0.32 0.21 0.99 0.000 0.41 0.19

P6 0.55 0.02 0.56 0.02 0.97 0.005 0.05 0.89

PD 0.12 0.66 na na na na 0.21 0.51

PspA1 0.09 0.74 na na na na 0.94 0.000

PspA2 0.82 0.000 0.83 0.000 0.27 0.67 0.7 0.011

Ply 0.06 0.98 na na Na na 0.35 0.27

CbpA 0.02 0.95 na na na na 0.29 0.36

a OM cohort only. b Both OM and control cohorts.

na: not available.

Linear regression analyses were performed for correlation.

Bold values represent strong correlation.

177

5.3.6 Serum and MEE cytokine titres in children with and without

RAOM/COME

Ten cytokines, including IL-1β, IL-5, IL-6, IL-8, IL-10, IL-12p70, IL-13, IL-17, IFN-γ and

TNF-α were measured in the serum and MEE from children with RAOM/COME and

serum of children without RAOM/COME. Cytokines were not detected in serum from

either group. Cytokines were detected in 0% (IL-5 and IL-13) to 94% (IL-8) of MEE

samples (Table 5-6). IL-1β was detected more often in those MEE which were positive

for bacterial otopathogens (Spn and/or Hi and/or Mcat) than in negative samples

(p=0.03; Table 5-6). TNF-α was only measureable in MEE samples positive for the

detection of Hi and/or Mcat (Table 5-6).

Table 5-6: MEE samples with cytokine concentrations which were higher than the limit

of detection (LOD) based on bacterial otopathogen presence in children from the OM2

cohort.

Cytokine Total MEE

(n=34)

MEE negative for bacterial

otopathogens

(n = 12)

MEE positive for bacterial

otopathogens

(n = 22)

p-value a

IL-1β 20 (59%) 4 (33%) 16 (73%) 0.03

IL-5 0 (0%) 0 (0%) 0 (0%) na

IL-6 24 (71%) 7 (58%) 17 (77%) 0.25

IL-8 32 (94%) 11 (92%) 21 (95%) 0.65

IL-10 23 (68%) 6 (50%) 17 (77%) 0.10

IL-12p70 3 (9%) 1 (8%) 2 (9%) 0.94

IL-13 0 (0%) 0 (0%) 0 (0%) na

IL-17 2 (6%) 1 (8%) 1 (5%) 0.65

IFN-γ 0 (0%) 0 (0%) 1 (5%) 0.45

TNF-α 5 (15%) 0 (0%) 5 (23%) 0.07

a p-value in comparison of frequencies of cytokine detection between MEE positive and negative for bacterial otopathogens using Pearson Chi-square.

178

5.3.6.1 Correlation of cytokine levels in MEE samples

To assess relationships amongst cytokines which could be produced from different

immune cells in regulation of innate immune responses against infection in the middle

ear, correlation of these cytokines were analysed either between left and right ears of

a same child or amongst cytokines in a same ear.

Levels of cytokines detected in left and right ears of a same child were investigated

for correlation. Only level of IL-8 showed strong correlation between left and right ears

of the same child (r=0.66, p=0.004).

Levels of 5 cytokines (IL-1β, IL-6, IL-8, IL-10 and TNF-α), which were detected in more

than 10% of MEE samples, were chosen for correlation analyses in the same ear.

Strong correlations were observed amongst levels of IL-1β, IL-10 and TNF-α as well

as between IL6 and IL8 (Table 5-7).

Table 5-7: Correlation in levels of cytokines in MEE samples from children with

RAOM/COME.

IL-6 IL-8 IL-10 TNF-α

IL-1β r=0.20

p=0.25

r=0.14

p=0.43

r=0.88

p=0.000

r=0.86

p=0.000

IL-6 r=0.38

p=0.03

r=0.29

p=0.10

r=0.005

p=0.98

IL-8 r=0.10

p=0.57

r=0.05

p=0.80

IL-10 r=0.77

p=0.000

Linear regressions were performed and the bolded r values represent significant correlation between levels of 2 cytokines present in the same ear with probability shown

179

5.3.6.2 Cytokine titres in relation to bacterial and viral pathogen presence in

the middle ear

The detection rates for bacterial and viral presence are reported in Chapters 3 and 4.

Those cytokines which were detected in more than 10% of MEE samples (IL-1β, IL-6,

IL-8, IL-10, TNF-α), were further analysed in the 34 MEE samples in relation to

pathogen presence. Interleukin 1-β, IL-6, IL-8 and IL-10 were significantly higher in

those MEE samples which were positive for any microorganism (bacterial

otopathogens and/or viruses), compared to negative samples (Figure 5-9a).

Interleukin 6 and IL-10, were higher when MEE samples were positive for bacterial

otopathogens (Spn and/or Hi and/or Mcat) than in negative samples (Figure 5-9b).

No differences were observed in cytokine levels between those MEE positive or

negative for viruses (Figure 5-9c).

Cytokine detection in relation to the presence and absence of the specific bacterial

(Spn, Hi and Mcat) and viral (specifically HRV) otopathogens were assessed to

determine if responses varied based on the species of pathogen in the middle ear. Of

the viral pathogens, only HRV was analysed in this manner as the remaining viruses

were not present in enough MEE to provide statistically comparable data (Chapter 3

and 4).

Titres of IL-8 were higher in those MEE samples positive for Hi than in negative

samples (p=0.008; Figure 5-10b). For those samples that were positive for Mcat, IL-

1β, IL-6 and TNF-α were significantly higher (p=0.01, p=0.002, p=0.001, respectively;

Figure 5-10c). No differences existed for any of the cytokines irrespective of Spn

detection (Figure 5-10a). For those MEE positive for HRV, IL-6 was significantly

higher (p=0.02; Figure 5-10d).

180

Cytokine levels vs viruses

pg

/mg

IL-1

bIL

-6IL

-8

IL-1

0

TNF-a

0.1

1

10

100

1000

10000

MEE positive for viruses

MEE negative for viruses

c

Cytokine levels vs bacteria

pg

/mg

IL-1

bIL

-6IL

-8

IL-1

0

TNF-a

0.1

1

10

100

1000

10000

MEE positive for bacteriaMEE negative for bacteria

p=0.03

p=0.04

b

Cytokine levels vs microbes

pg

/mg

IL-1

bIL

-6IL

-8

IL-1

0

TNF-a

0.01

0.1

1

10

100

1000

10000

MEE positive for mircobes

MEE negative for microbes

p=0.05

p=0.01

p=0.02

p=0.01

a

Figure 5-9: Cytokines in MEE based on microbial detection. Comparisons were

performed using Student’s t tests on log-transformed data. Values represent GMC with

95% CI.

181

Cytokine levels vs Spnp

g/m

g

IL-1

bIL

-6IL

-8

IL-1

0

TNF-a

0.01

0.1

1

10

100

1000

10000

MEE positive for SpnMEE negative for Spn

a Cytokine levels vs Hi

pg

/mg

IL-1

bIL

-6IL

-8

IL-1

0

TNF-a

0.1

1

10

100

1000

10000

MEE positive for HiMEE negative for Hi

p=0.008

b

Cytokine levels vs Mcat

pg

/mg

IL-1

bIL

-6IL

-8

IL-1

0

TNF-a

0.1

1

10

100

1000

10000

MEE positive for Mcat

MEE negative for Mcat

p=0.01

p=0.002

p=0.001

cCytokine levels vs HRV

pg

/mg

IL-1

bIL

-6IL

-8

IL-1

0

TNF-a

0.1

1

10

100

1000

10000MEE positive for HRV

MEE negative for HRVp=0.02

d

Figure 5-10: Levels of cytokines in MEE positive and negative for specific bacteria and virus. Data were analysed using Student’s t

tests on log-transformed data. Values represent GMC with 95% CI.

182

5.3.7 Regulatory T cells in peripheral blood mononuclear cells in


T lymphocyte subsets, including CD4+ T cells, CD4+FoxP3 T cells, CD4+CD25+

T cells and CD4+CD25+FoxP3 T cells (Treg cells), were investigated within

PBMC to assess frequency differences between children with and without

RAOM/COME. Only those samples with a cell viability of over 90% were analysed

to avoid non-specificity in antibody binding to dead cells. The average viability of

PBMC from both groups was 95% 2% (median 95%; range from 90% to 98%).

No significant differences in proportions of T lymphocyte subsets were observed

between children with and without RAOM/COME (Table 5-8).

Table 5-8: Proportion of T lymphocyte subsets in PBMC of children from the OM

and control cohorts.


(n = 12)

Children without

RAOM/COME

(n = 12)

p-value

% CD4+ (in lymphocyte gate) 44.6 7.5 40.6 6.7 0.21

% CD4+FoxP3+ (in CD4+ gate) 0.3 0.2 0.3 0.1 0.98

% CD4+CD25+ (in CD4+ gate) 5.4 2.2 5.6 2.0 0.83

% CD4+CD25hi 1.9 1.1 2.1 1.0 0.75

% CD4+CD25low 3.5 1.4 3.6 1.3 0.92

% CD4+CD25+FoxP3+ (in CD4+CD25+ gate)

11.2 6.6 9.8 5.0 0.59

% CD4+CD25hiFoxP3+ 9.1 5.9 8.1 4.4 0.67

% CD4+CD25lowFoxP3+ 2.1 1.3 1.7 0.9 0.37

Values represent mean with standard deviation. Student’s t tests were performed for comparison.

183

5.4 Discussion

This study investigated serum, salivary and MEE IgG and IgA against 4 protein

vaccine candidates for Spn (PspA1, PspA2, Ply and CbpA) and 4 for Hi (P26, P4,

P6 and PD) in young children with and without history of RAOM/COME. We

demonstrated that children with RAOM/COME had similar or higher antibody

titres to antigens of Spn and Hi, compared with children without RAOM/COME.

These results are consistent with findings from other studies showing that

children with RAOM/COME had a normal capacity to produce serum antibodies

against antigens of Spn and Hi (Misbah et al., 1997, Menon et al., 2012,

Wiertsema et al., 2012, Corscadden et al., 2013). In contrast, other studies have

indicated that children with RAOM/COME have lower serum IgG, but not IgA,

against proteins of Spn and Hi, compared to children without RAOM/COME (Kaur

et al., 2011a, Kaur et al., 2011b). The differences in methodology for antibody

detection, case-definition of otitis prone, and the time of sample collection make

it difficult to directly compare results between studies. Serum samples in the

studies by Kaur et al. were collected at the onset of an AOM episode, compared

to the current study in which serum was collected at the time of VTI surgery when

the child did not have an acute infection. While pathogen specific serum IgG titres

were not significantly different between children with MEE positive or negative for

Spn or Hi, consistent with the study by Wiertsema et al. (Wiertsema et al., 2012),

IgA to pneumococcal proteins were higher in the serum of children whose MEE

was positive for Spn in the current study. Colonisation with Spn and Hi was

associated with increased levels of antigen specific IgA and IgG in serum of

children with RAOM/COME. Wiertsema et al. have previously reported that

specific IgG to Spn and Hi proteins were higher in children who were colonised

with these otopathogens compared to children who were not (Wiertsema et al.,

2012). Together these results support the conclusion that children with

RAOM/COME have a normal ability to produce antibodies against bacteria.

Spn and Hi specific antibodies were frequently detected in the MEE children with

RAOM/COME despite relatively few being positive for Spn (12%) or Hi (26%) by

PCR (Chapter 3). These data may suggest that antibody responses may

contribute to the elimination of the bacteria from the middle ear or that

184

colonisation induces a widespread mucosal response. Previous studies have

demonstrated that antibody in the middle ear is associated with clearance of

clinical infection in children with AOM (Howie et al., 1973, Sloyer et al., 1976,

Karjalainen et al., 1991b, Karjalainen et al., 1991a). We also demonstrated that

anti-Spn and Hi protein antibody titres were higher in MEE which were negative

for bacteria than those which were positive, suggesting that MEE antibodies may

have cleared these infections.

When the relationship between sampling sites of the same children were

assessed, IgG showed a stronger correlation between left and right ears, as well

as between MEE and serum than IgA, suggesting that IgA is locally produced at

the specific mucosal induction site whilst IgG is present at the mucosal sites via

transudations. Indeed, previous studies have demonstrated that the middle ear

is a major site for local IgA production (Howie et al., 1973, Sloyer et al., 1975,

Jones et al., 1979). Collectively, children with RAOM/COME showed a similar

capacity to produce both local and systemic antibody as their control counterparts

with a suggestion that these may have the capacity to eradicate bacteria from the

middle ear.

IgA, against antigens of Spn and Hi was present in all saliva samples and at

similar titres from children regardless of their OM history. In contrast, the

presence of IgG in saliva varied depending on antigens. These results are

consistent with findings from other studies, where anti-pneumococcal

polysaccharides IgA presented frequently in saliva of children in comparison to

rare detection of IgG (Simell et al., 2002, Simell et al., 2007).

A major limitation of this study is the small number of saliva samples that were

collected at a high enough volume for analysis. Despite this, we found

significantly higher levels of anti-P26 IgG in saliva of children with RAOM/COME

than in children with no significant OM history, suggesting that antibody may be

produced in response to Hi infection so as to neutralise this pathogen. In addition,

strong correlations in titres of salivary, MEE and serum anti-P6 IgG, suggesting

that it is systemically produced and present at the mucosal surfaces via a process

of transudation. Furthermore, strong correlations in levels of antibodies,

particularly IgA, against Spn and Hi antigens between saliva and serum,

185

suggesting that these saliva antibodies may be systemic products. Increased

levels of salivary anti-pneumococcal protein antibodies have been associated

with a lower risk of AOM development (Simell et al., 2007, Xu et al., 2015),

demonstrating important roles of mucosal antibodies in prevention of OM

development.

Cytokines play important roles in immune protection from bacterial and viral

infections (Imanishi, 2000). Indeed, systemic cytokines, such as IL-1β, IL-6, IL-8,

IL-10, IL-13 and TNF-α, have been shown to be elevated in children with bacterial

(Scharer et al., 2003) or viral (Patel et al., 2009) AOM. In the current study, we

did not detect any of the cytokines measured in the serum of children with or

without RAOM/COME. The lack of systemic cytokine detection in the current

study may be due to the collection of serum from “well” children compared to

previous studies where serum was collected at the time of acute

infection/inflammation (Scharer et al., 2003, Patel et al., 2009).

Cytokines, particularly IL-8, were frequently detected in the MEE from children

with RAOM/COME. These results are consistent with previous studies where IL-

1β, IL-6, IL-8 and IL-10 were detected in the MEEs of children with RAOM/COME

(Maxwell et al., 1994, Skotnicka and Hassmann, 2000, Zielnik-Jurkiewicz and

Stankiewicz-Szymczak, 2015). We also demonstrated that significantly higher

levels of cytokines were present in those MEE which were positive for any

microbes than in negative samples. Our results are consistent with the study by

Stol et al., where elevated MEE cytokines were observed when the sample was

positive for bacteria (Stol et al., 2012). The presence of Hi and Mcat in particular

appear to lead to increased cytokines in the middle ear. In contrast, viral detection

was not associated with an increase in cytokine in the middle ear in either our

study or Stol’s previous study (Stol et al., 2012) with the exception being for HRV

which was associated with an increased IL-6 titre. This induction of IL-6 which is

a pro-inflammatory cytokine may suggest an important role of this virus in OM

pathogenesis however this requires further investigation.

Treg cells (CD4+CD25+FoxP3 T cells) are hypothesised to perform an important

role in pathogenesis of recurrent and chronic OM due to their ability to regulate

and, in particular, diminish immune responses to microbial infection (Zhang et al.,

186

2011). Previously, adults with COME were demonstrated to have a significantly

higher percentage of Treg cells in their blood, compared to healthy controls (Zhao

et al., 2009). This study demonstrated that the percentages of T cell subsets

within PBMC, including CD4+CD25+ T cells and CD4+CD25+FoxP3+ T cells,

were the same for children with or without RAOM/COME. These results are

consistent with findings from previous studies showing that children with OME

and adenoidal hypertrophy had the same proportions of CD4+CD25+ T cells in

both blood and adenoids, compared to those with adenoidal hypertrophy alone

(Kotowski et al., 2011). A limitation of both our and the reports by Kotowski et al.

are that the control groups were not “healthy” as they were undergoing

adenoidectomy for adenoidal hypertrophy, which may in turn, be caused by

chronic infections or allergy (Rout et al., 2013, Scadding, 2010, Al-Mazrou and

Al-Khattaf, 2008, Modrzynski and Zawisza, 2007). In addition, the current study

demonstrated that all adenoids samples, regardless of the child’s OM history,

were positive for at least one of the 3 predominant bacteria (Chapter 3) and/or

viral pathogen (Chapter 4). Multi-microbe infections were also common, being

detected in more than 40% of the adenoid samples, suggesting that adenoids

from the children of both the OM and control cohorts may be in a chronic state of

infection. This further evidence reinforces the issue of availability and selection

of relevant control cohorts and emphasises that children with adenoidal

hypertrophy may not be “ideal” controls for OM studies, including cellular

investigations. Furthermore, frozen PBMC for Treg cells analyses may also be a

limitation in the current study due to reduced cells viability and viable cell recovery

compared to fresh PBMC as described in Chapter 2. Significant reduces of CD4+

T and CD4+CD25+ T cell populations in frozen PBMC compared to fresh cells

were reported (Costantini et al., 2003, Elkord, 2009). That may impact on Treg

cells analyses as they are a very rare population in human PBMC (Koristka et al.,

2014). Collectively, “real” healthy controls and fresh PBMC should be considered

for future studies of OM with Treg cells.

This study demonstrated that children with RAOM/COME do not exhibit an

impaired immune response which supports the development and implementation

of new pneumococcal or NTHi vaccines containing protein candidates to help

prevent OM pathogenesis in both OM-prone and non-prone children.

187

Chapter 6:

Systemic immune gene expression in

urban children with and without

RAOM/COME in South East Queensland

188

6.1 Introduction

The immune system provides the first line of defence against colonisation and

subsequent infection by disease causing pathogens. A rapid, appropriate and

regulated response is vital to eliminate any infection that occurs. Expression of

the relevant genes is the first step to modulate the immune responses and

eradicate infection. Dysregulation of immune gene expression may result in

differential inflammatory responses to infectious pathogens and the development

of chronic and recurrent OM (Liu et al., 2013a).

To investigate the role of gene expression on development and progression of

AOM, both animal models and humans have been investigated. In rat models,

genes encoding a robust inflammatory response were upregulated during Spn

infection in the animals (Chen et al., 2005). Changes in gene expression underpin

a potential balance between positive and negative regulation of the inflammatory

process during AOM induced by NTHi in the murine model (Hernandez et al.,

2015). The upregulation of NLRP3 and ASC genes that lead to pro-IL1β and pro-

IL-18 processing occur within 24h in NTHi infected animals compared to controls

(Hernandez et al., 2015). Together with upregulation of genes related with pro-

inflammation, upregulation of genes related to anti-inflammatory responses, such

as IRAK3 and NLRP12, were also observed in the NTHi infected animals within

24h and remained until 120h, compared to the controls (Hernandez et al., 2015).

Changes in gene expression and its relevance in regulation of particularly anti-

inflammatory responses subsequent to initial infection are significant in the

overall control of pathogenesis of OM in the middle ear (Hernandez et al., 2015).

Regulation of the host immune response to bacterial infection of the middle ear

in humans can include both up-regulation and down-regulation of gene

expression. Significant up-regulation of genes has been associated with up-

regulation of innate immune bacterial defences, such as DEB123, S100A12 and

S100A9, whilst down-regulation of genes related to B-cells activation and

proliferation, such as CD19, BLNK and FOXP1, was observed in patients

experiencing Spn induced AOM (Liu et al., 2012). In contrast, down regulation of

gene expression was associated with stimulation of pro-inflammatory cytokine

responses, such as IL6 and CCL2, during NTHi induced AOM in children (Liu et

189

al., 2013a). Clearly, differential responses in gene expression are associated with

regulation of host immune responses that may limit or control colonisation and

infection due to a range of bacterial otopathogens, including the predominant Spn

and NTHi during AOM.

While studies investigating gene regulation profiles in AOM have been

conducted, there is no data on profiling the immune gene expression of children

with recurrent or chronic OM exists. It is possible that the chronic or recurrent

nature of the disease in otitis prone children may result from a deficiency in the

host’s innate and/or adaptive immune responses to the pathogen, or that

dysfunction or imbalances in immune gene regulation within the host may

contribute to ongoing inflammation and inappropriate control of the host response

and control of infections (Wiertsema and Leach, 2009). In this study, the potential

role of immune gene regulation of host immune response to otopathogens was

investigated in children with and without RAOM/COME. Furthermore, gene

expression were compared between children with RAOM/COME, who were MEE

positive and negative for bacterial otopathogens or who were MEE positive and

negative for any viruses. Nanostring technology was used to assess the

expression of 579 genes related to immune responses in peripheral blood

mononuclear cells (PBMC) of children with RAOM/COME and compared these

to controls with no significant history of OM.

6.2 Methods

Patient recruitment, sample collection and processing, PBMC isolation and

Nanostring gene expression assays were performed as described in Chapter 2,

sections 1, 2, 13 and 15, respectively.

Raw Nanostring gene expression data were normalised using negative controls,

positive controls, and housekeeping genes via nSolver version 2.6 software

(Nanostring). The maximum value of the 8 internal negative controls in each

sample was subtracted from the gene expression count to exclude nonspecific

detection. The geometric mean of the 6 internal positive controls was used to

normalise the data so as to minimise falsification from batch effects and to ensure

that samples could be compared. Gene expressions were then normalised using

190

the geometric mean of the 15 housekeeping genes (ABCF1, ALAS1, EEF1G,

G6PD, GAPDH, GUSB, HPRT1, OAZ1, POLR1B, POLR2A, PPIA, SDHA, TBP,

TUBB and RPL19), which are expressed consistently in PBMC.

Student’s t tests were performed automatically by nSolver to compare means of

individual gene expression between children with RAOM/COME and control

groups, between MEE positive and negative for bacterial otopathogens and

between MEE positive and negative for any viruses. MEE samples were positive

or negative for the presence of bacterial otopathogens (Spn, Hi or Mcat) as

determined by PCR (Chapter 3) as well as, and scored either positive or negative

for the presence of any of the tested viruses in the middle ear (ADV, hMPV, IAV,

IBV, RSV, HRV, PIV1, PIV2, PIV3, WU, HboV, 229E, HKU-1, NL63, OC43, EV

and KI, (Chapter 4).

Normalised data for gene expression and fold change were uploaded into

Ingenuity Pathway Analysis (IPA) software (Ingenuity® Systems,

www.ingenuity.com) to analyse immune response pathways, functions and

networks. Each gene was mapped to its corresponding gene object in the

Ingenuity Pathways Knowledge Base. The information in the Ingenuity

Knowledge Base (Genes only) was used as a reference set for both direct and

indirect relationships. The data sources from ingenuity expert findings and the

“Core Analysis” function were used to interpret the data in context of biological

functions, pathways and networks. Genes with 10 or more copies and which were

differentially expressed with a fold change of 1.5 or greater or -1.5 or less were

overlaid onto global molecular networks, which are developed from information

contained in the ingenuity knowledge base. Networks were then algorithmically

created based on their connectivity. Networks were “named” on the most

prevalent functional groups present. Canonical pathways analysis identified the

most significant reference pathways in the IPA database that fit the datasets that

were entered.

Host and environmental risk factors were compared between children from the

OM2 and control cohorts using Student’s t tests for continuous variables (age and

number of episodes of OM) and Pearson Chi-Square analyses (p-value

http://www.ingenuity.com/

191

asymptotic significant 2-sided) for categorical variables (gender, exposure to

cigarette smoke, allergy and asthma).

Some individual genes may be expressed differentially in children with asthma

and/or allergy, for example, there is an association between asthma and the level

of gene expression of TLR4 and TNF-α in children under 6 years of age (Klaassen

et al., 2015) whilst children with a food allergy have reduced levels of FoxP3 and

IL-10 gene expression than observed in healthy children (Krogulska et al., 2011).

In this study, the difference in the number of children with asthma and/or allergy

was investigated before comparing differential levels of gene expression between

the 2 groups of children. The IBM SPSS Statistics 23 for Windows software

package (IBM, New York, USA) was used for all statistical analyses. Data were

plotted using GraphPad Prism 5.0 (Graph Software Inc., California, USA).

192

6.3 Results

6.3.1 Study populations

In this analysis, 9 children with RAOM/COME, who were identified as otitis prone

(OP), and 7 children without a significant history of RAOM/COME, who were

identified as non-otitis prone (NOP) children, were analysed for gene expression

(Table 6-1). Children were identified as OP children if they experienced up to 4

or more episodes of AOM within 12 months or middle ear effusion lasting longer

than 3 months. In contrast, NOP children had experienced fewer than 3 episodes

of AOM within 12 month period. Two OP children were reported to be allergic by

their parents (1 to cow’s milk protein and 1 to penicillin) (Table 6-1). One NOP

child was allergic to gluten (Table 6-1). Two NOP children had asthma (Table 6-

1) and no significant differences in the number of children having asthma and/or

allergy was observed between the 2 groups of children (Table 6-1).

Table 6-1: Demographic data for children in each cohort.

OP children NOP children p-value

Number 9 7

Mean age in year (range) 2.8 (2–4) 3.9 (3–5) 0.02

Male (%) 6 (67%) 6 (89%) 0.38


6.4 (4-10) 1.0 (0-2) 0.000

Exposed to cigarette smoke 3 (33%) 1 (14%) 0.38

Any allergy a 2b (22%) 1c (14%) 0.92

Asthma 0 (0%) 2 (28%) 0.08

a Cow’s milk protein, egg, wheat, shellfish, house dust mite, pollens/trees, moulds, grasses, bee/insect venom, peanut, other nuts, cat/dog hair, others.

b One child allergic to cow’s milk protein and 1 allergic to penicillin.

c One child allergic to gluten.

Student’s t tests were used to compare continuous variables (age and number episodes of OM).

Pearson Chi-Square analyses were used to compare categorical variables (male, exposure to cigarette smoke, any allergy and asthma).

193

6.3.2 Gene expression between OP and NOP children

Genes, which had 10 or more copies following normalisation with the nSolver

software, were considered to be expressed. There were no significant differences

in the numbers of genes expressed in PBMC between OP and NOP children

amongst the 594 genes that were analysed, including the 15 housekeeping genes

(417 vs 418). Due to the small sample size of the study, cut-offs for fold change

values 1.5 or ≤ -1.5 were used to compare differences in expression levels of

examined genes between OP and NOP children regardless p-value. Sixteen

genes were expressed differentially between 2 groups of children (Table 6-2).

More specifically, six genes were expressed at higher rates and 10 genes were

expressed at lower rates in OP children compared with NOP children (Table 6-

2). Notably, 75% of OP children versus 43% of NOP children had genes

expressed encoding for HLA-DQA1 and HLA-DQB1 in PBMC; however the

difference was not statistically significant. HLA-DQA1 and HLA-DQB1 expression

were 11.6 and 7.6 fold higher in OP compared to NOP children, respectively

(Table 6-2).

194

Table 6-2: Annotated genes which were differentially expressed (fold change

1.5 or ≤ -1.5) in PBMC between OP and NOP children.

Gene symbol

Gene title Fold change

(OP/NOP)

p-value

HLA-DQA1 Major histocompatibility complex, class II, DQ alpha 1

11.630 0.16

HLA-DQB1 Major histocompatibility complex, class II, DQ beta 1

7.560 0.19

HLA-DRB1 Major histocompatibility complex, class II, DR beta 1

1.840 0.56

NT5E 5'-nucleotidase, ecto (CD73) 1.760 0.37

CD9 CD9 molecule 1.570 0.57

CR2 Complement component receptor 2 1.530 0.44

KIR3DL2 Killer cell immunoglobulin-like receptor, three domains, long cytoplasmic tail, 2

-2.210 0.07

SERPING1 Serpin peptidase inhibitor, clade G (C1 inhibitor), member 1

-1.960 0.34

TNFRSF8 Tumour necrosis factor receptor superfamily, member 8

-1.900 0.19

CCR5 Chemokine (C-C motif) receptor 5 (gene/pseudogene)

-1.790 0.21

IL15 Interleukin 15 -1.720 0.22

C2 Complement component 2 -1.700 0.20

CFD Complement factor D (adipsin) -1.660 0.14

CARD9 caspase recruitment domain family, member 9

-1.570 0.12

C6 Complement component 6 -1.520 0.26

PPBP Pro-platelet basic protein (chemokine (C-X-C motif) ligand 7)

-1.510 0.53

Student’s t tests were performed automatically by nSolver software to compare means of individual gene expression between 2 groups.

195


Differential gene expression between OP and NOP children was used to identify

and predict canonical or established host response pathways. There were 35

canonical pathways observed, based on the fold changes of gene expression

between OP and NOP children. One dendritic cell maturation pathway

significantly predicted “activation” or “upregulation” in OP children (Figure 6-1).

No “inhibition” or “downregulation” pathways were determined in OP children.

These findings suggest that immune responses were activated in these children.

196

Figure 6-1: Canonical pathways predicted in OP children.

The right-tailed Fisher Exact test was used to calculate p-value. These probabilities were calculated by considering the number of focus genes

that participate in a process and the total number of genes known to be involved in that process in the selected reference set in IPA software.

The ratio is the number of genes that meet the cut-off criteria / the number of genes that make up the pathway. The cut-off criteria were that a

gene in one group was counted 10 copies and had the fold change 1.5 or ≤ -1.5 compared with the same gene in another group.

Positive/negative z-score is considered as activated/inhibited state of the process, respectively. Positive z-score (orange colour) represents

activated or upregulated state of a pathway whilst negative z-score (blue colour) represents inhibited or downregulated state of a pathway.

197

6.3.2.2 Biological functions and networks of gene expression in OP

children

The biological functions and networks of identified genes were examined. One

biological function was predicted to be significantly different in OP children based

on analysis of the identified genes. The function encoded by these genes showed

a decrease in “immune cell death” in OP children compared to NOP children

(activation z-score = 2.108, p = 4.67E-6) (Figure 6-2). Identified genes in this

biological function included CCR5, IL15, TNFRSF8 and PPBP, which are all

associated with immune cell death. These genes had lower levels of expression

in OP children compared to NOP children (Figure 6-2), whilst CR2, which is

associated with inhibition of cell death of immune cells, was, in contrast, more

highly expressed in OP children (Figure 6-2).

Figure 6-2: Shown is the decreased “cell death of immune cells” function

predicted in OP children.

Leads to inhibition

Gene expressed higher in OP children

Gene expressed lower in OP children

Predicted decrease or inhibition

198

For the data set generated in this study, biological networks are algorithmically

generated pathways describing potential molecular interactions for a given

dataset. In this study, five biological networks were predicted by the algorithm,

based on differential gene expression between the 2 groups of children assessed

(Table 6-3). Only one network (#1) involved the majority (10/16) of genes

identified above with high score (>10) (Table 6-3), the score indicates the

likelihood that the assembly of a set of focus genes in a network could be

explained by random chance alone.

199

Table 6-3: Predicted networks relating differentially expressed genes between

OP and NOP children.

No. Networks Genes expressed

higher in OP children

Genes expressed lower in OP

children

Score Focus molecules

1 Cell-To-Cell signalling and interaction, Haematological system development and function, Immune cell trafficking

CCL8, CD244, CD40LG, CR2, EBI3, HLA-DQA1, HLA-DQB1, HLA-DRB1, IL1RN, MAPK1, NT5E, PSMB8, TLR1, TNF, TP53, TRAF1

CARD9, CFD, IL15, IL27, PPBP, PSMP10, RELA, TNFRSF8

23 10

2 Developmental disorder, Hereditary disorder, Immunological disease

GRAPH C2 3 1

3 Cardiovascular disease, Organismal injury and abnormalities, Haematological disease

- SERPING1 3 1

4 Cellular Movement, Haematological system development and function, Immune cell trafficking

CCL5, CCL7, CCL8, CCL13, CD4, IFNG, IFNL2, IL14, TLR2, ZAP70

CCL2, CCL3, CCL4, CCR5, CEBPB, HAVCR2,

2 1

5 Cancer, Organismal injury and abnormalities, Cellular movement

CD9, CD36, CD46, CD81, ITGA6

CD82, ITGB1, KIT

1 1

The genes found to be differentially expressed in our experiments and the number of such genes displayed in the “Focus molecules” column have been highlighted in bold italic that meet the criteria cut-off and/or filter criteria, and were mapped to its corresponding gene object in IPA Knowledge base.

The score is generated using a p-value calculation. This score indicates the likelihood that the assembly of a set of focus genes in a network could be explained by random chance alone. The data base attributed general cellular functions to each network which are determined by interrogating the Ingenuity Pathway Knowledge base for relationships between the genes in the network and the cellular functions they impact.

200

Figure 6-3: Network #1 showed connection amongst identified genes. Green

shapes indicate those genes that were expressed at a lower rate in OP children,

compared to NOP children. Red shapes indicate those genes more highly

expressed in OP children than in NOP children. White and grey shapes indicate

genes involved in the networks but the fold changes of expression between OP

and NOP did not meet the cut-off setting (1.5 or ≤-1.5).

201

6.3.3 Differences in gene expression amongst OP children with

MEE positive and negative for bacterial otopathogens

Gene expression within the host immune response may differ in association with

the presence or absence of bacterial otopathogens (Spn, Hi and Mcat) within the

middle ear. Gene expression within OP children, whose MEE were positive or

negative for bacterial otopathogens were investigated. Six OP children had MEE

which were PCR positive for bacterial otopathogens (MEE-POS-BAC) and 3

children had MEE which were PCR negative for bacterial otopathogens (MEE-

NEG-BAC) in this participant subset selected for gene expression analyses.

A total of 84 genes were differentially expressed between children who had MEE

that were PCR positive or negative for bacterial otopathogens. Among the 84

genes, 65 showed increased expression whilst 19 genes were less frequently

expressed, in MEE-POS-BAC, compared to MEE-NEG-BAC samples (Table 6-

4).

Innate immune response genes, such as MSR1 and MARCO (encoding

receptors expressed on macrophages), S100A8/9 (encoding calcium-binding

proteins which are highly expressed in neutrophil and monocyte cytosol) and

genes encoding Toll-like receptors, were expressed with a fold-change greater

than 1.5 in children with bacteria detected in the middle ear compared to those

without (Table 6-4). In contrast, adaptive immune response genes, such as

CD40LG (encoding a receptor expressed on activated T cells) and CCR6

(encoding a chemokine receptor expressed on dendritic cells, B and T cells),

showed reduced expression with a fold change of -1.5 or lower in children with

bacteria detected in the middle ear compared to those without (Table 6-4).

202

Table 6-4: Genes that were differentially expressed (fold change 1.5 or ≤ -1.5)

in PBMC between children whose MEE was positive or negative for any of the

main bacterial otopathogens.

Gene symbol

Gene title Fold change (MEE-POS-

BAC vs MEE-NEG-BAC)

p-value

MSR1 Macrophage scavenger receptor 1 7.270 0.21

TNFRSF8

Tumour necrosis factor receptor superfamily, member 8 6.220

0.21

SERPING1

Serpin peptidase inhibitor, clade G (C1 inhibitor), member 1 5.010

0.06

MARCO Macrophage receptor with collagenous structure 4.180

0.14

C2 Complement component 2 4.000 0.19

LILRA3

Leukocyte immunoglobulin-like receptor, subfamily A (without TM domain), member 3 3.960

0.39

TLR5 Toll-like receptor 5 3.760 0.22

CCR2 Chemokine (C-C motif) receptor 2 3.240 0.13

IL15 Interleukin 15 3.030 0.40

HLA-DRB1

Major histocompatibility complex, class II, DR beta 1 2.680

0.24

IFIT2 Interferon-induced protein with tetratricopeptide repeats 2 2.650

0.37

MX1 MX dynamin-like GTPase 1 2.500 0.009

CFD Complement factor D (adipsin) 2.440 0.30

BATF3 Basic leucine zipper transcription factor, ATF-like 3 2.250

0.33

LILRB4

Leukocyte immunoglobulin-like receptor, subfamily B (with TM and ITIM domains), member 4 2.250

0.21

GBP1 Guanylate binding protein 1, interferon-inducible 2.170

0.03

CLEC4E C-type lectin domain family 4, member E 2.140 0.17

CARD9 Caspase recruitment domain family, member 9 2.120

0.15

HLA-DQA1

Major histocompatibility complex, class II, DQ alpha 1 2.030

0.79

TNFSF10 Tumour necrosis factor (ligand) superfamily, member 10 2.030

0.004

CXCL10 Chemokine (C-X-C motif) ligand 10 2.020 0.29

CX3CR1 Chemokine (C-X3-C motif) receptor 1 1.990 0.10

CSF1R Colony stimulating factor 1 receptor 1.940 0.25

LTBR Lymphotoxin beta receptor (TNFR superfamily, member 3) 1.900

0.01

203

IRAK3 Interleukin-1 receptor-associated kinase 3 1.890

0.08

LY96 Lymphocyte antigen 96 1.880 0.14

CD1D CD1d molecule 1.870 0.10

FCER1G Fc fragment of IgE, high affinity I, receptor for; gamma Polypeptide 1.860

0.01

EGR1 Early growth response 1 1.820 0.06

LILRB3


0.24


S100A8 S100 calcium binding protein A8 1.780 0.08

LILRA5 Leukocyte immunoglobulin-like receptor, subfamily A (with TM domain), member 5 1.760

0.19



0.11




TNFSF13B

Tumour necrosis factor (ligand) superfamily, member 13b 1.710

0.04

CCL3 Chemokine (C-C motif) ligand 3 1.700 0.15

LILRB1


0.02

S100A9 S100 calcium binding protein A9 1.690 0.09

KIR3DL2 Killer cell immunoglobulin-like receptor, three domains, long Cytoplasmic tail, 2 1.680

0.65

CD36 CD36 molecule (thrombospondin receptor) 1.660

0.10

FCGR2A Fc fragment of IgG, low affinity IIa, receptor (CD32) 1.640

0.15

IFNAR1 Interferon (alpha, beta and omega) receptor 1 1.640

0.000

LILRB2


0.20

NOD2 Nucleotide-binding oligomerization domain containing 2 1.620

0.09

IL8 Interleukin 8 1.600 0.11

IFI35 Interferon-induced protein 35 1.600 0.08

BST2 Bone marrow stromal cell antigen 2 1.580 0.02

IRF7 Interferon regulatory factor 7 1.580 0.05

CSF2RB

Colony stimulating factor 2 receptor, beta, low-affinity (granulocyte-macrophage) 1.570

0.12

CYBB Cytochrome b-245, beta polypeptide 1.560 0.08

204


0.11

PDGFRB Platelet-derived growth factor receptor, beta polypeptide 1.550

0.72

STAT2 Signal transducer and activator of transcription 2, 113kDa 1.550

0.01

KLRC1 Killer cell lectin-like receptor subfamily C, member 1 1.540

0.40

NCF4 Neutrophil cytosolic factor 4, 40kDa 1.540 0.008

PRKCD Protein kinase C, delta 1.540 0.04


IRF5 Interferon regulatory factor 5 1.530 0.09

PYCARD PYD and CARD domain containing 1.530 0.21

HAVCR2 Hepatitis A virus cellular receptor 2 1.510 0.22

IL1B Interleukin 1, beta 1.510 0.27

CD79A CD79a molecule, immunoglobulin-associated alpha -1.520

0.27

CD22 CD22 molecule -1.530 0.36

ICOSLG/LOC102723996 Inducible T-cell co-stimulator ligand -1.530

0.13

NFKB1 Nuclear factor of kappa light polypeptide gene enhancer in B-cells 1 -1.550

0.17

TNFRSF13C

Tumour necrosis factor receptor superfamily, member 13C -1.560

0.29

IFNG Interferon, gamma -1.570 0.24

S1PR1 Sphingosine-1-phosphate receptor 1 -1.570 0.21

IL32 Interleukin 32 -1.590 0.21

PAX5 Paired box 5 -1.600 0.35

CD40LG CD40 ligand -1.610 0.14

IL21R Interleukin 21 receptor -1.630 0.10

TRAF4 TNF receptor-associated factor 4 -1.630 0.16

HLA-DOB

Major histocompatibility complex, class II, DO beta -1.730

0.32

CCR6 Chemokine (C-C motif) receptor 6 -1.780 0.40

CR2 Complement component (3d/Epstein Barr virus) receptor 2 -1.790

0.31

LTA Lymphotoxin alpha -1.820 0.08

LAG3 Lymphocyte-activation gene 3 -1.860 0.23

PPBP Pro-platelet basic protein (chemokine (C-X-C motif) ligand 7) -1.910

0.53

CD9 CD9 molecule -2.790 0.11


205

6.3.3.1 Canonical pathways of gene expression in OP children with

MEE positive for bacterial otopathogens

Differential expression of genes (Table 6-4) associated with the presence of

bacteria in the middle ear of OP children led to the prediction of 50 canonical

pathways. Of these 50 pathways, 11 were predicted to be associated with

“activation” or “upregulation” and six were predicted to relate to “inhibition” or

“downregulation” of genes expressed in children whose MEE were positive for

bacterial otopathogens (Table 6-5).

Given the nature of canonical pathways as being specific to a type of immune

response, the “inhibition” or “downregulation” of a pathway in the MEE-POS-BAC

group would mean “activation” or “upregulation” in the MEE-NEG-BAC group.

The majority of pathways related to innate immune responses were more active

in MEE-POS-BAC whilst pathways related to adaptive immune responses were

more active in MEE-NEG-BAC (Table 6-5). For examples, “Toll-like receptors

signalling” pathway was activated in MEE-POS-BAC whilst “iCOS-iCOSL

signalling in T helper cells” pathway was activated in MEE-NEG-BAC (Table 6-

5).

206

Table 6-5: Canonical pathways predicted in MEE-POS-BAC children.

Canonical pathways Prediction Z-score p-value

Toll-like receptors signalling Activation 3.000 7.42E-14

Role of pattern recognition receptors (PRR) in recognition of bacteria and viruses

Activation 2.496 1.82E-20

TREM1 signalling Activation 2.309 9.98E-16

Activation of IRF by cytosolic PRR Activation 1.633 5.37E-07

Complement system Activation 1.342 2.03E-08

Interferon signalling Activation 1.342 9.39E-07

LPS-stimulated MAPK signalling Activation 1.342 3.23E-05

NF-κB signalling Activation 1.291 5.46E-15

Dendritic cell maturation Activation 1.147 2.90E-22

Production of Nitric Oxide (NO) and reactive oxygen species in macrophages


MIF regulation of innate immunity Activation 1.000 5.16E-5

iCOS-iCOSL signalling in T helper cells Inhibition -2.000 4.67E-08

CD40 signalling Inhibition -2.000 2.97E-04

Acute myeloid leukemia signalling Inhibition -2.000 6.31E-04

LXR/RXR activation Inhibition -1.134 1.64E-07

Lymphotoxin β receptor signalling Inhibition -1.342 7.33E-06

B cell receptor signalling Inhibition -1.342 2.41E-05

The right-tailed Fisher Exact Test was used for statistical analyses and generation of p-value. The probability was calculated by considering the number of focus genes that participate in a process and the total number of genes that are known to be involved in that process in the selected reference set in IPA software. Positive/negative z-score was considered as activated/inhibited state of the process, respectively.

207


children with MEE positive for bacterial otopathogens

Twenty nine immune functions related to the gene expression dataset, were

predicted to be increased when children had bacteria detected in their MEE (ie

MEE-POS-BAC) (Table 6-6). No functions were predicted to decrease in these

children. Genes which may lead to an enhanced immune response in children

with bacterial detection from their MEE are illustrated in Figure 6-4. These

include, but are not limited to, activation of innate immune responses, such as

“macrophage differentiation” (Figure 6-4a), “production of reactive oxygen

species” (Figure 5-4b) and “cell binding” (Figure 6-4c).

Table 6-6: Biological functions of immune responses predicted in children

who had MEE that were positive for bacterial otopathogens.

Diseases or Functions Annotation

Predicted Activation State

z-score p-value

Binding of cells Increased 2.898 1.09E-15

Adhesion of blood cells Increased 2.896 1.25E-22

Adhesion of Tumour cell lines Increased 2.779 2.64E-05

Expression of RNA Increased 2.708 1.19E-05

Proliferation of smooth muscle cells

Increased 2.595 1.77E-06

Transcription Increased 2.523 1.22E-05

Synthesis of reactive oxygen species


Differentiation of macrophages Increased 2.379 1.04E-08

Cell death Increased 2.293 2.60E-18

Binding of professional phagocytic cells


Proliferation of vascular smooth muscle cells


Engulfment of cells Increased 2.223 2.29E-07

Production of reactive oxygen species (ROS)


Inflammatory response of cells Increased 2.219 1.27E-09

Generation of ROS Increased 2.219 1.88E-05

208

Response of endothelial cells Increased 2.213 1.27E-09

Shape change of blood cells Increased 2.207 1.08E-06

Adhesion of leukemia cell lines Increased 2.200 8.85E-05

Cell movement of T lymphocytes Increased 2.194 7.17E-06

Internalization of cells Increased 2.155 4.14E-09

Differentiation of mononuclear leukocytes


Migration of mononuclear leukocytes


Cell movement of lymphocytes Increased 2.046 1.64E-07

Lymphocyte migration Increased 2.045 7.70E-07

Adhesion of immune cells Increased 2.043 2.57E-21

Cell movement of carcinoma cell lines


Binding of blood cells Increased 2.001 4.81E-11

Necrosis Increased 2.000 1.61E-14

Inflammatory response of endothelial cells


The right-tailed Fisher Exact Test was used for statistical analyses and generation of p-value. The probability was calculated by considering the number of focus genes that participate in a function as well as the total number of genes that are known to be involved in that function in the selected reference set in IPA software. Positive/negative z-score is considered as increased/decreased activation of the function, respectively.

209

Figure 6-4: Three biological functions predicted in children with bacterial

otopathogens detected from their MEE. a) genes involved in increased function

of “differentiation of macrophages”; b) genes involved in increased function of

“production of reactive oxygen species”; c) genes involved in increased function

of “binding of cells”.

Leads to activation.

Predicted activation. The colour intensity indicates the confidence of prediction.

Gene expressed lower in MEE-POS-BAC. The colour intensity indicates the expression of gene.

Gene expressed higher in MEE-POS-BAC. The colour intensity indicates the expression of gene.

210

Eight biological networks were predicted based on differential gene

expression in OP children due to the presence of bacterial otopathogens in

the middle ear (Table 6-7). Only 4 of these 8 networks were predicted with

a high score (>10) (Table 6-7) and related to immune responses targeting

microbial infection in those children whose MEE were positive for

otopathogens.

211

Table 6-7: Predicted networks relating to differentially expressed genes between OP children who had MEE which were positive or

negative for bacterial otopathogens.

No. Networks Genes expressed higher in MEE-POS-BAC

Genes expressed lower in MEE-POS-BAC


1 Immunological Disease, Infectious Diseases, Cell-To-Cell Signalling and Interaction

CSF2RB, CYBB, GBP1, HLA-DRB1, IFI35, IFIT2, IFNAR1, IRF5, IRF7, KLRC1, MX1, STAT2, TLR7, TNFSF10, TNFSF13B

CCR6, CD22, CD79A, CR2, IFNG, NFKB1, PAX5, TNFRSF13C

43 23

2 Cell-To-Cell Signalling and Interaction, Haematological System Development and Function, Immune Cell Trafficking

CARD9, CCL3, CCR2, CD14, CD36, CD86, CD1D, FCGR2A, LILRA2, LTBR, LY96, MSR1, PYCARD, TLR2, TLR4, TLR8, TNFRSF8

CD9, CD40LG, ICOSLG/LOC102723996, IL21R, LTA,

41 22

3 Haematological System Development and Function, Inflammatory Response, Tissue Morphology

CX3CR1, CXCL8, EGR1, FCER1G, HAVCR2, IL15, IL1B, LILRA5, LILRB1, LILRB4, NCF4, NOD2, PRKCD, S100A8, S100A9, TLR5

IL32, S1PR1 31 18

4 Cancer, Organismal Injury and Abnormalities, Cellular Movement

BST2, CCL7, , CD14, CD163, CFD, CLEC4E, CYBB, FCER1G, HLA-DQA1, IFNGR1, IRAK3, LILRA3, LILRB2, MARCO

CCL11, CCL18, HLA-DOB, IL13, , IL27, IRAK2, MRC1, PPBP, TNF, TRAF4

22 14

5 Cellular Movement, Haematological System Development and Function, Immune Cell Trafficking

C2, CSF1R, CSF2RA, PDGFRB, TLR2, TNFSF10

NFIL3, PDGFB, TGFBR1, TNFSF11

7 6

212

6 Cell Death and Survival, Haematological System Development and Function, Haematopoiesis

- LAG3 2 1

7 DNA Replication, Recombination, and Repair, Gene Expression, Molecular Transport

BATF3 - 2 1

8 Cardiovascular Disease, Organismal Injury and Abnormalities, Digestive System Development and Function

SERPING1 - 2 1

The genes found to be differentially expressed in our experiments and the number of such genes displayed in the “Focus molecules” column have been highlighted in bold italic that meet the criteria cut-off and/or filter criteria, and were mapped to its corresponding gene object in IPA Knowledge base.

The score is generated using a p-value calculation. This score indicates the likelihood that the assembly of a set of focus genes in a network could be explained by random chance alone. The data base attributed general cellular functions to each network which are determined by interrogating the Ingenuity Pathway Knowledge base for relationships between the genes in the network and the cellular functions they impact.

213

6.3.4 Differences in gene expression amongst OP children with

MEE positive and negative for any tested virus

Differences in gene expression were examined in OP children, whose MEEs were

positive or negative for any tested virus. MEE samples for 6 children were positive

for the presence of at least one of the viruses assessed in the middle ear (MEE-

POS-V). Three were negative for all viruses assessed in the middle ear (MEE-

NEG-V) by PCR.

Fifty five genes were differentially expressed between the 2 groups of children.

Forty five of these genes showed an enhanced expression and 10 showed a

reduced expression in MEE-POS-V compared to MEE-NEG-V (Table 6-8). Of the

55 genes which were differentially expressed in this analysis, 37 genes were also

previously observed between MEE-POS-BAC and MEE-NEG-BAC, including

MSR1, MARCO and CCR6 (Table 6-8).

214

Table 6-8: Genes which are differentially expressed (fold change 1.5 or ≤ -1.5)

in PBMC between OP children with positive or negative MEE for viruses tested.

Gene symbol

Gene title Fold Change (MEE-POS-V/ MEE-NEG-V)

p-value

MSR1* Macrophage scavenger receptor 1 11.740 0.08

C1QB Complement component 1, q subcomponent, B chain

7.160 0.001

TNFRSF8*

Tumour necrosis factor receptor superfamily, member 8

5.660 0.24

KIR3DL1 Killer cell immunoglobulin-like receptor, three domains, long cytoplasmic tail, 1

4.610 0.05

C2* Complement component 2 4.490 0.15

CXCL1 Chemokine (C-X-C motif) ligand 1 (melanoma growth stimulating activity, alpha)

4.420 0.05

MARCO* Macrophage receptor with collagenous structure

4.390 0.12

C1QA Complement component 1, q subcomponent, A chain

4.140 0.03

IL15* Interleukin 15 4.030 0.27

LILRA3* Leukocyte immunoglobulin-like receptor, subfamily A (without TM domain), member 3

4.020 0.39

CLEC6A C-type lectin domain family 6, member A

3.790 0.04

TLR5* Toll-like receptor 5 3.170 0.31

CFD* Complement factor D (adipsin) 3.100 0.14

HLA-DRB1*

Major histocompatibility complex, class II, DR beta 1

3.010 0.17

CD55 CD55 molecule, decay accelerating factor for complement (Cromer blood group)

2.870 0.27

HLA-DQB1

Major histocompatibility complex, class II, DQ beta 1

2.570 0.69

HLA-DQA1*

Major histocompatibility complex, class II, DQ alpha 1

2.410 0.75

BATF3* Basic leucine zipper transcription factor, ATF-like 3

2.370 0.29

215

SERPING1*

Serpin peptidase inhibitor, clade G (C1 inhibitor), member 1

2.130 0.49


LILRB4* Leukocyte immunoglobulin-like receptor, subfamily B (with TM and ITIM domains), member 4

1.940 0.35

CSF1R* Colony stimulating factor 1 receptor 1.900 0.27

HAVCR2*

Hepatitis A virus cellular receptor 2 1.850 0.01

CXCL10* Chemokine (C-X-C motif) ligand 10 1.830 0.38

NLRP3 NLR family, pyrin domain containing 3 1.830 0.005

CLEC4E* C-type lectin domain family 4, member E

1.800 0.33

IL1B* Interleukin 1, beta 1.790 0.12


TNFRSF17

Tumour necrosis factor receptor superfamily, member 17

1.760 0.32

EGR2 Early growth response 2 1.740 0.09

IRAK3* Interleukin-1 receptor-associated kinase 3

1.730 0.18

IL1R2 Interleukin 1 receptor, type II 1.690 0.26

LILRA5* Leukocyte immunoglobulin-like receptor, subfamily A (with TM domain), member 5

1.690 0.25

LY96* Lymphocyte antigen 96 1.670 0.25

KIR3DL2*

Killer cell immunoglobulin-like receptor, three domains, long cytoplasmic tail, 2

1.610 0.67

CARD9* Caspase recruitment domain family, member 9

1.600 0.44

CD1D* CD1d molecule 1.580 0.31

GBP1* Guanylate binding protein 1, interferon-inducible

1.580 0.27

IL8* Interleukin 8 1.570 0.13

CCRL2 Chemokine (C-C motif) receptor-like 2 1.550 0.30



1.530 0.45

216

ATG12 Autophagy related 12 1.520 0.45


1.510 0.34

CXCL2 Chemokine (C-X-C motif) ligand 2 1.500 0.23

CD160 CD160 molecule -1.500 0.14

CR2* Complement component (3d/Epstein Barr virus) receptor 2

-1.500 0.52

HLA-DOB*

Major histocompatibility complex, class II, DO beta

-1.510 0.49

TNFRSF10C

Tumour necrosis factor receptor superfamily, member 10c, decoy without an intracellular domain

-1.510 0.68

IL32* Interleukin 32 -1.570 0.23

BLNK B-cell linker -1.600 0.48

CCR6* Chemokine (C-C motif) receptor 6 -1.600 0.50

IFNG* Interferon, gamma -1.630 0.19

CD9* CD9 molecule -1.660 0.5

LAG3* Lymphocyte-activation gene 3 -1.780 0.26

*Gene also differentially expressed in the previous analysis between MEE-POS-BAC and MEE-NEG-BAC.


6.3.4.1 Canonical pathways of gene expression in OP children with

MEE positive for any tested viruses

Similarly to the comparison between MEE-POS-BAC and MEE-NEG-BAC, more

than 50 canonical pathways were predicted from the dataset of differential gene

expression based on viral presence in the middle ear of OP children. Amongst

these pathways, 9 were predicted to be “activated” or “upregulated” and one was

predicted to be “inhibited” or “downregulated” in the MEE-POS-V (Table 6-9). The

activated immune response pathways were mostly related to innate immune

response pathways, including the “role of PRR in recognition of bacteria and

217

viruses”, “dendritic cell maturation” and “TLR signalling”. The observed activation

of TLR signalling is illustrated in Figure 6-5 to demonstrate the genes involved.

Table 6-9: Predicted canonical pathways in OP children with MEE positive for

any of the viruses analysed.

Canonical pathways Prediction Z-score p-value

Role of PRR in recognition of bacteria and viruses


Dendritic cell maturation Activation 1.667 1.10E-09

NF-κB signalling Activation 2.121 2.01E-08

TLRs signalling Activation 2.449 4.88E-08

TREM1 signalling Activation 2.449 5.30E-08

iNOS signalling Activation 1.000 6.10E-06

PKCθ signalling in T lymphocytes Activation 1.000 2.97E-04

IL-8 signalling Activation 1.000 1.57E-03

LPS/IL-1 mediated inhibition of RXR function


LXR/RXR activation Inhibition -2.000 1.92E-05

The right-tailed Fisher Exact Test was used for statistical analyses and generation of p-value. The probability was calculated by considering the number of focus genes that participate in a process and the total number of genes that are known to be involved in that process in the selected reference set in IPA software. Positive/negative z-score is considered as activated/inhibited state of the process, respectively.

218

Figure 6-5: The TLR

signalling pathway (an innate

immune response pathway)

was predicted to be activated

in OP children who had virus

positive MEE. The pink or

purple shapes indicated that

genes were more highly

expressed in PBMC from

children whose MEE was

positive for any virus

compared to those with

negative MEE. No genes

showed a reduced expression

in those children with virus

positive MEE compared to

those with negative MEE in

this pathway.

219


children with MEE positive for any tested virus

There were 59 functional immune responses predicted from the gene expression

dataset. Fifty six functions were predicted to be “increased” and 3 functions were

predicted to be “decreased” in those children who had MEE which were positive

for any of the viruses assessed (Table 6-10). Majority of “upregulated” or

“increased” expression of genes in MEE-POS-V related to increase immune

responses essential to eliminate microbial infection and decrease morbidity,

mortality or infection. Two predicted functions (one “increased” and one

“decreased”) are illustrated in Figure 6-6 to demonstrate specific genes involved.

220

Table 6-10: Biological functions of immune responses predicted in PBMC from

OP children with MEE that were positive for any of the viruses assessed.

Diseases or Functions Annotation

Predicted Activation

State

Activation z-score

p-value

T cell development Increased 3.415 1.70E-13

Immune response of cells Increased 3.348 1.26E-21

Generation of cells Increased 3.168 1.04E-10

T cell homeostasis Increased 3.144 1.78E-14

Lymphocyte homeostasis Increased 3.122 1.98E-16

Development of leukocytes Increased 3.063 8.63E-18

Cellular homeostasis Increased 2.910 7.06E-12

Development of lymphocytes Increased 2.875 4.92E-15

Aggregation of cells Increased 2.766 1.20E-05

Binding of blood cells Increased 2.763 2.41E-08

Differentiation of T lymphocytes Increased 2.665 2.54E-11

Cell movement Increased 2.586 2.30E-16

Migration of phagocytes Increased 2.586 2.05E-08

Aggregation of blood cells Increased 2.580 9.43E-06

Recruitment of phagocytes Increased 2.551 9.55E-13

Recruitment of cells Increased 2.546 3.21E-18

Adhesion of blood cells Increased 2.473 3.23E-19

Quantity of leukocytes Increased 2.457 1.21E-18

Leukocyte migration Increased 2.443 6.60E-23

Accumulation of neutrophils Increased 2.430 1.27E-09

Differentiation of mononuclear leukocytes


Recruitment of leukocytes Increased 2.404 1.22E-17

Recruitment of neutrophils Increased 2.369 3.82E-13

Leukocytosis Increased 2.349 1.25E-09

Response of mononuclear leukocytes


Migration of cells Increased 2.297 6.37E-17

Quantity of cells Increased 2.290 5.86E-17

221

Quantity of mononuclear leukocytes


Allergic inflammation Increased 2.236 3.98E-09

Delayed hypersensitive reaction


Immune response of leukocytes


Migration of neutrophils Increased 2.230 2.74E-08

Differentiation of leukocytes Increased 2.208 3.10E-17

Quantity of blood cells Increased 2.207 8.88E-21

Proliferation of lymphatic system cells


Differentiation of helper T lymphocytes


Recruitment of antigen presenting cells


Attraction of mononuclear leukocytes


Cellular infiltration by lymphocytes


Cell-mediated response Increased 2.122 2.38E-11

Differentiation of Th17 cells Increased 2.122 1.04E-10

Quantity of T lymphocytes Increased 2.103 2.97E-07

Accumulation of granulocytes Increased 2.070 1.96E-11

Attraction of leukocytes Increased 2.066 9.48E-10

Chemoattraction Increased 2.066 2.40E-09

Cell movement of T lymphocytes


Innate immune response Increased 2.047 1.97E-11

Arthritis Increased 2.044 4.33E-22

Activation of antigen presenting cells


Response of lymphocytes Increased 2.030 3.63E-14

Accumulation of leukocytes Increased 2.029 2.36E-12

Synthesis of reactive oxygen species


Quantity of lymphocytes Increased 2.016 2.23E-13

222

Immune response of T lymphocytes


Recruitment of myeloid cells Increased 2.012 3.46E-16

Cell movement of leukocytes Increased 2.005 4.68E-17

Morbidity or mortality Decreased -2.565 7.26E-09

Infection of mammalia Decreased -2.711 1.12E-21

Organismal death Decreased -2.763 5.60E-07

The right-tailed Fisher Exact Test was used for statistical analyses and generation of p-value. The probability was calculated by considering the number of focus genes that participate in a function and the total number of genes that are known to be involved in that function in the selected reference set in IPA software. Positive/negative z-score is considered as increased/decreased activation of the function, respectively.

223

Figure 6-6: “Upregulated” or “increased” expression of genes involved in increase of “development of lymphocytes” and in decrease

of “infection of mammalia” in children with MEE that was positive for any of the viruses assessed. Genes CD1D, TLR4, MSR1, TLR5,

IL15, TLR7, IL1B and CLE6A are involved in both functions. This data was in comparison between MEE-POS-V and MEE-NEG-V.

Gene expressed higher in MEE-POS-V. The colour intensity indicates the expression of gene.

Predicted activation. The colour intensity indicates the confidence of prediction.

Predicted inhibition. The colour intensity indicates the confidence of prediction.

Leads to activation.

Leads to inhibition

224

Seven biological networks were predicted based on differential gene expression

in OP children due to the presence of viruses in the middle ear (Table 6-11),

however, only 2 networks had scores more than 10 (Table 6-11). These 2

networks related to inflammatory responses, and cellular maintenance in immune

responses to eradicate viral infection in those children who had virus positive

MEE (Table 6-11).

225

Table 6-11: Predicted networks relating to differentially expressed genes between children who had MEE that were positive or

negative for any of the viruses assessed.

No. Networks Genes expressed higher in MEE-POS-V

Genes expressed lower in MEE-POS-V


1 Inflammatory Response, Cell-To-Cell Signalling and Interaction, Cellular Function and Maintenance

CARD9, CD1D, CXCL8, GBP1, HAVCR2, HLA-DQB1, IL15, IL1B, IRAK3, LILRB4, LY96, MSR1, NLRP3, TLR4, TLR5, TLR7, TNFRSF8, TNFRSF17

CCR6, IFNG, IL32, 45 21

2 Cellular Function and Maintenance, Cellular Development, Cellular Growth and Proliferation

ATG12, CDKN1A, CFD, CSF1R, HAVCR2, HLA-DQA1, HLA-DRB1, ITGA2B, LILRA3, LILRA5, LY96, TLR5, TNFRSF8

CD9, CD244, HLA-DOB, IL27, IL21R, TNF, TRAF1,

26 14

3 Cell-To-Cell Signalling and Interaction, Haematological System Development and Function, Immune Cell Trafficking

C3, CCL3, CCRL2, CD36, CD86, CLEC4E, CTNNB1, CXCL8, FCER1G, IL10, IL1B, LILRB2, STAT1, STAT3

CCL2, CCL22, IL13, RELA, TNF

8 5

4 Cancer, Organismal Injury and Abnormalities, Cell Death and Survival

C2, EGR2, GAPDH, MARCO, PDGFRB

BLNK, CD6, PDGFB, TGFBR1

7 5

5 Cell Death and Survival, Haematological System Development and Function, Haematopoiesis

- LAG3 2 1

6 DNA Replication, Recombination, and Repair,

BATF3 - 2 1

226

Gene Expression, Molecular Transport

7 Cardiovascular Disease, Organismal Injury and Abnormalities, Digestive System Development and Function

SERPING1 - 2 1

The genes found to be differentially expressed in our experiments and the numbers of these genes are displayed in the “Focus molecules” column. Highlighted genes (bold italic) met the criteria cut-off and/or filter criteria and were mapped to their corresponding gene object in IPA Knowledge base.

The score is generated using a p-value calculation and indicates the likelihood that the assembly of a set of focus genes in a network could be explained by random chance alone. The data base attributed general cellular functions to each network, which are determined by interrogating the Ingenuity Pathway Knowledge base for relationships between the genes in the network and the cellular functions they impact.

227

6.4 Discussion

This is the first study to investigate immune gene expression profiles in OP and

NOP children using Nanostring technology. Overall, systemic immune gene

expression in OP Children was similar to that of NOP children. OP children

however, expressed higher levels of genes which were related to dendritic cell

maturation, a key process that links the host’s innate and adaptive immune

responses (Banchereau and Steinman, 1998). Recently, Surendran et al.,

reported that OP children had significantly higher number of monocytes and

conventional dendritic cells compared to NOP children (Surendran et al., 2016).

That study also revealed similar levels of secretion of 11 innate

cytokine/chemokines (IL-1β; IL-6, IL-8, IL-10, IL-12p70, TNFα, IFN-γ, CCL2,

CCL4, CCL5 and CXCL10) investigated during this study. Furthermore, similar

expression levels of TLR2, TLR4, TLR7, TLR8, MyD88, TRIF, IRAK3, IRF3 and

IRF7 in PBMC from OP children compared with NOP children were noted

(Surendran et al., 2016). In the current study, gene expression levels of cytokine

and TLR signalling pathways from PBMC were also shown to be similar between

OP and NOP children (supplemental Figure 6-1). These results suggest that

innate immune responses are not deficient in OP children and furthermore,

results from Chapter 5 of the results in this thesis demonstrate that OP children

have similar or higher serum antibody titres in response to antigens from Spn and

Hi, compared with non OP children. Previous research including the current

Chapter 5 of this study and previous reports (Wiertsema et al., 2012, Menon et

al., 2012, Corscadden et al., 2013) reported that OP children have similar or

higher serum antibody titres to antigens from Spn and Hi identified within NOP

children. Together, these findings demonstrate that OP children have normal

systemic immune responses at both gene and protein levels.

All children with RAOM/COME in this study had effusion in the middle ear at the

time of enrolment, however not all MEE tested positive for bacterial otopathogens

and/or viruses which were analysed using PCR. This may indicate that the

pathogens in those MEE had already been eradicated by the host immune

response. We demonstrated that genes related to innate immune responses

were more highly expressed in those children who had a pathogen detected in

228

their middle ear, whilst genes related to adaptive immune responses were more

highly expressed in children with MEE negative for microbes. For example,

S100A8 and S100A9, which are highly expressed in neutrophils from murine and

chinchillas infected by NTHi (Hernandez et al., 2015, Kerschner et al., 2009),

were significantly more highly expressed in PBMC from children who had one of

the bacterial otopathogens detected in their MEE, compared to those where the

MEE was negative. In contrast, CD40LG, which is preferentially expressed by T

and B cells (Wang et al., 2001, Kawabe et al., 2011), was more highly expressed

in children with MEE that were negative for bacteria (supplemental Figure 6-2).

In addition, our results (Chapter 5) and results from Stol et al. (Stol et al., 2012)

have demonstrated significantly higher levels of cytokines, including IL-1β and IL-

8, in MEE which are positive for bacterial otopathogens compared with to those

MEE which are negative. In contrast, antibody titres to antigens of Spn and Hi

were observed to be higher in MEE which were negative for bacteria compared

with those which were positive (Chapter 5). Combined with the gene data, these

findings suggest that children with otopathogen infection within the middle ear,

have innate immune responses that are activated. In contrast, adaptive immune

responses were activated in those children with no otopathogens detected,

perhaps after the innate immune mechanisms had cleared the original infection.

Notably, we demonstrated in this small study, that OP children expressed the

HLA-DQA1 and HLA-DQB1 genes more often than NOP children (75% vs 43%).

These genes are associated with celiac disease, a congenital chronic

inflammatory condition of the small intestine (Dubois et al., 2010). Although an

association between the expression of HLA-A2 and RAOM has been reported

before (Kalm et al., 1991, Kalm et al., 1994), no study has investigated expression

of HLA-DQA1 and HLA-DQB1 genes in any disease including OM. Our pilot data

suggests that levels of expression, mutations and allelic variations of these genes

should be assessed in children with a history of RAOM/COME in a larger study

population.

A wide range of immuno-genes was investigated in both OP and NOP children in

this study. While subject numbers were small, the innate immune response

pathway was clearly more highly activated in OP children compared to NOP

229

children. Detection of otopathogens in the middle ears of OP children was

associated with activation of innate immune response pathways, whilst adaptive

immune response pathways appear to be more highly activated in those OP

children which were negative for otopathogens. Expression of HLA-DQA1 and

HLA-DQB1 genes are different between OP and NOP children and should be

investigated in RAOM/COME pathogenesis. Larger longitudinal studies should

be conducted to investigate potential roles of the genes and pathways identified

in OM development of chronic or recurrent diseases. Overall, Nanostring

technology has enhanced understanding of the complex interaction of regulation

of the innate and adaptive immune responses in OP and NOP children.

230

Cytokine gene expression in PBMC between OP and NOP chichildren

Co

un

t

IL1-

bIL

-5IL

-6IL

-8

IL-1

0

IL-1

2A

IL-1

2BIL

-13

IL-1

7

IFN-g

TNF

CCL2

CCL4

CCL5

CXCL10

0.1

1

10

100

1000

10000

OP

NOP

TLR gene expression in PBMC between OP and NOP chichildren

Co

un

t

TLR1

TLR2

TLR3

TLR4

TLR5

TLR7

TLR8

TLR9

1

10

100

1000

OP

NOP


of cytokines and TLRs in PBMC of OP and NOP children. Comparisons of levels

gene expression between OM and control cohorts were analysed using Student’s

t tests on log-transformed data. Values represent geometric mean with 95% CI.

231

Co

un

t

IL-1

bIL

-8

TLR1

TLR2

S10

0A8

S10

0A9

CD40

LG

CD79

A

1

10

100

1000

10000

MEE-POS-BAC MEE-NEG-BAC

**

**


in PBMC of OP children with MEE positive and negative for bacteria.

Comparisons of levels gene expression between OM and control cohorts were

analysed using Student’s t tests on log-transformed data. Values represent

geometric mean with 95% CI. Asterisk indicates significant with p<0.05.

232

Chapter 7:

General discussion

233

More than 80% of children will experience otitis media (OM) prior to their third

birthday, and, as one of the most common childhood infections, OM has been

suggested to be an inevitable illness of childhood (Rovers et al., 2004).

Approximately 70% of children living in developed countries, will recover

spontaneously from acute episodes of OM within 7 days of onset, whilst the

remaining children will experience either chronic OM, characterised by

persistence of effusion in the middle ear for more than three months, or recurrent

episodes of acute OM (RAOM)(Bluestone, 2003, Rosenfeld and Kay, 2003). In

addition to the significant impact of OM on an individual child and their family, the

treatment or failure to treat and prevent OM places a significant burden on the

wider community and health care system. Effective OM prevention and treatment

is a priority for reducing hearing loss and subsequent sequelae, worldwide.

Childhood RAOM and COME have long been investigated to define the

multifactorial pathogenesis in otitis prone children. These efforts are essential to

enable development of treatments or preventative strategies that will stop the

complications that arise from this disease. Principally, ET dysfunction, microbial

infection and the host immune response are the key factors for OM pathogenesis,

however these all have complex interactions with multiple environmental factors

(Rovers et al., 2004, Massa et al., 2015). The complicated interactions between

microbial infection of the middle ear and the host immune response to infection

make it difficult to clearly identify whether otitis prone children are susceptible to

microbial infection due to environmental risk factors or if they have compromised

immune responses.

Bacteria have a major role in OM pathogenesis, particularly the predominant

bacterial otopathogens S. pneumoniae, H. influenzae and M. catarrhalis (Massa

et al., 2009). Globally, S. pneumoniae has been identified as the predominant

pathogen of AOM whilst H. influenzae has been detected as the most common

bacterial otopathogen of RAOM/COME (Ngo et al., 2016). M. catarrhalis follows

these and is the third most commonly detected otopathogenic bacteria from the

middle ear (Ngo et al., 2016).

234

This study demonstrates that within South East Queensland, H. influenzae

remains the predominant bacteria detected in the URT, including the middle ear,

of children with RAOM/COME. More specifically, all H. influenzae strains

detected in cohorts recruited in 2015 were NTHi, which is consistent with the

results of previous reports from Western Australia (Wiertsema et al., 2011b), and

the Netherlands (Stol et al., 2013). Furthermore, in our study, the frequency of

beta-lactamase producing H. influenzae increased significantly between the pilot

cohort (OM1) recruited between Dec 2008 and May 2010, and the 2015 cohorts

(OM2 and control) to a higher level than observed in previous reports (Wiertsema

et al., 2011b, Pumarola et al., 2013). This increased capacity for bacterial

resistance to beta lactam antibiotics highlights the need for ongoing surveillance.

Elsewhere in Australia, particularly in remote communities housing Indigenous

Australian children, the frequency of NTHi induced OM was reduced by the use

of a 10-valent pneumococcal polysaccharide conjugate vaccine (Synflorix;

Glaxo-Smith Kline), which contains the protein D antigen from NTHi as a carrier

protein (Leach et al., 2015). Prevention of OM using a vaccine approach is

dependent upon children being able to mount an effective antibody response.

Results from the current study confirmed that children with RAOM/COME were

able to produce local and systemic IgG and IgA against the protein antigens P4,

P6, P26 and PD of H. influenzae, as did children with AH or OSA, the control

cohort. Thus, these results suggest that a vaccine based on these antigens would

induce an efficacious immune response that may prevent H. influenzae

infection/colonisation and potentially reduce OM caused by this bacterium.

In the present study the frequency of S. pneumoniae detection within the URT of

children was observed to reduce slightly between the 2 cohorts recruited in 2008-

2011 and 2015. Introduction of PCV13 into the Australian National Immunisation

Programme in 2011 (Johnson et al., 2012) may have had a role in reducing the

prevalence of S. pneumoniae in the URT, including the middle ear, in these

children. Reductions in the frequency of pneumococcal detection from the URT

have been reported for many countries since the introduction of PCV7 into their

respective NIPs (Casey et al., 2010, Dupont et al., 2010, Hoberman et al., 2011,

Alonso et al., 2013). Despite a transient fall in the frequency of pneumococcal

detection in the URT after the introduction of PCV7, this was subsequently

235

reported to increase due to serotype replacement with non-vaccine serotypes,

particularly 19A (Casey et al., 2010, Fenoll et al., 2011). Subsequently, PCV13,

which includes 7 serotypes of PCV7 and 6 additional serotypes, including 19A,

has been introduced into the NIP of many developed countries, including the

United States and Australia (Kaplan et al., 2013, Johnson et al., 2012).

Serotype 19A detection from the middle ears of children with OM was reduced

following the introduction of PCV-13 immunisation. In contrast, non-PCV-13

vaccine serotypes including 35B, 23A/B, 15 and 11 are more frequently detected

in otitis prone children (Kaplan et al., 2015). Results from our study indicate that

the frequency of pneumococcal detection in the middle ear of children with OM

has decreased after PCV13 introduction. Overall, non-PCV13 serotypes were

identified in 69% (11/16) and 47% (7/15) of pneumococcal isolates in the URT

from children with and without RAOM/COME, respectively. These results are

consistent with other studies in countries where PCV13 was introduced into the

NIP (Ben-Shimol et al., 2014, Ben-Shimol et al., 2016, Martin et al., 2014a,

Angoulvant et al., 2015, Kaplan et al., 2015, Ochoa-Gondar et al., 2015, Desai et

al., 2015) and highlight concerns that PCV7 may no longer be effective in

preventing pneumococcal infection/colonisation in the near future, when non-

PCV13 pneumococcal serotypes become predominant. While it is unlikely that a

vaccine against more than 90 serotypes of S. pneumoniae (Habib et al., 2014)

can be developed or is feasible, a vaccine based on protein antigens, such as

PspA, CbpA and Ply that present commonly in all pneumococcal serotypes

(Darrieux et al., 2015), may be more successful in preventing OM due to S.

pneumoniae. Furthermore, our current results indicate that children with and

without RAOM/COME are able to produce both local and systemic IgA and IgG

against these protein antigens. These results suggest that otitis prone children

are likely to respond to a protein-based pneumococcal vaccine and that this may

induce an efficacious response to prevent pneumococcal infection/colonisation

and OM development due to S. pneumoniae.

The frequency of M. catarrhalis detection in the middle ear was observed to

increase over the period of patient recruitment in the current study. This trend

towards an increased M. catarrhalis detection frequency followed PCV13

236

introduction, although the number of patients recruited in the current study was

not sufficient for statistical confirmation. These results are consistent with a recent

report from the United States, in which M. catarrhalis was found to be the

predominant bacteria detected in children with OM after PCV13 introduction

(Casey et al., 2015b). AOM due to M. catarrhalis has been indicated to be

associated with lower frequencies of tympanic membrane perforations, increased

mixed infections and younger age at first diagnosis, compared with AOM caused

by other bacterial otopathogens (Broides et al., 2009). M. catarrhalis is estimated

to contribute to up to 142 million cases of AOM globally each year in both children

and adults (Monasta et al., 2012). Of significance, 100% of M. catarrhalis isolates

in children with OM are reported as being β-lactamase producers and are able to

resist to β-lactam antibiotics, including Amoxicillin, which is the recommended

first-line antibiotic to treat children with AOM. Thus, OM treatment is complicated

by the need to combine the antibiotic and a β-lactamase inhibitor, such as

Amoxicillin and a clavulanic acid (β-lactamase inhibitor) (Drawz and Bonomo,

2010, Atkinson et al., 2015). Currently, an effective vaccine for M. catarrhalis is

not available although several vaccine candidates are under investigation

(Ruckdeschel et al., 2009, Ren et al., 2015, Ren and Pichichero, 2016).

In the present study, associations between the 3 main bacterial otopathogens

identified within different URT locations of the same children were investigated.

Overall, the majority of bacteria detected within the middle ear were also detected

in the nasopharynx and/or adenoids of the same child with RAOM/COME. These

results support previous reports identifying associations between

nasopharyngeal colonisation with bacterial otopathogens and development of

OM in young children (Faden et al., 1997, Pettigrew et al., 2011, Ruohola et al.,

2013). Furthermore, in previous studies, the bacterial otopathogens within the

middle ear and adenoids were seen to be significantly associated and to be

genetically similar (Emaneini et al., 2013). More specifically, within the current

study, only S. pneumoniae detected in the middle ear was also detected in

nasopharynx and/or adenoids, which is consistent with previous studies.

Interestingly, detection of H. influenzae and M. catarrhalis in the middle ear was

not always associated with detection of these bacteria in other locations of the

URT. This finding is consistent with previous study that reported only 80% of H.

237

influenzae strains detected in the middle ear were genetically similar to strains

found in the nasopharynx of a same child with RAOM/COME (Stol et al., 2013).

These results suggest that H. influenzae and M. catarrhalis may persist in the

middle ear of children with RAOM/COME in the form of bacterial biofilms or

intracellular infections to maximise survival. Furthermore, reinfection of the

middle ear may not be reliant upon direct seeding of bacterial otopathogens from

the URT, specifically those located within the nasopharynx and/or adenoids of a

same child. Bacterial biofilms and intracellular infection and survival of these

bacteria have been reported as a key mechanism for recurrent and chronic OM

(Hall-Stoodley et al., 2006, Thornton et al., 2011, Thornton et al., 2013).

Overall, up to 60% of H. influenzae isolated from the middle ear during a recurrent

AOM episode were reported as genetically similar to strains isolated at an initial

AOM episode of a same child. Thus, approximately two-thirds of H. influenzae

may persist in the middle ear from the initial infection through biofilms and then

reinfect the middle ear subsequently (Kaur et al., 2013). Collectively, these results

suggest that bacteria causing the infection at the first episode of AOM, are more

likely to be seeded directly from the nasopharynx and/or adenoids of a same

child. Subsequent episodes of OM in the same child may result from a

combination of bacterial persistence in biofilms and/or intracellular infection

facilitating bacterial survival. Whether otitis prone children are more susceptible

to biofilm formation and thus at greater risk of recurrent or chronic OM is

unknown. It is postulated that antibiotics could potentially stimulate planktonic

bacteria to form the complex biofilm structure to protect themselves from

antibacterial agents (Wu et al., 2014).

In conjunction with bacterial otopathogens, the current study identified a range of

viruses in the URT of children with and without RAOM/COME. Otitis media can

be caused by viruses alone, although more commonly, bacteria are also present

within the middle ear (Heikkinen and Chonmaitree, 2000, Pettigrew et al., 2011,

Nokso-Koivisto et al., 2015). Globally, a range of viruses have been identified as

causal for OM, including, respiratory syncytial virus, adenovirus or rhinovirus,

which are reported to be more frequently identified in the middle ear of children

with OM (Shaw et al., 1995, Chonmaitree et al., 2012, Wiertsema et al., 2011a,

238

Yano et al., 2009b). Rhinovirus was the predominant virus detected in all

locations of the URT of children in the current study and this is consistent with

previous reports which also identify this as being the most common virus detected

in children with OM (Chantzi et al., 2006, Wiertsema et al., 2011a, Stol et al.,

2013).

The presence of rhinovirus within the URT was also associated with increased

cytokine responses, such as IL-1β and IL-6, in the middle ear of children with

RAOM/COME (Chapter 5) and provides further evidence of this virus’s potential

role in facilitation of OM development. Our study also highlighted the association

between rhinovirus detection with the middle ear and concurrent detections within

the nasopharynx and/or adenoids of the same child. This finding provides

evidence for the potential involvement of reservoirs for viral pathogens in the

development of OM.

Recently identified viruses such WU polyomavirus and bocavirus, were included

in the multiplex PCR panels used in the current study. This is one of only a few

studies that report on a wide range of viruses in the URT of children with OM

(Wiertsema et al., 2011a). Indeed, the current study, is, to our knowledge, the

first report comparing a wide range of viruses within a range of URT locations

including the middle ear and adenoids from children with RAOM/COME or

AH/OSA. All children were positive for at least one tested virus in the adenoids

and the majority of adenoids examined had multiple viruses identified. These

results support a hypothesis of viral flora in the adenoids of children, particularly

those with enlarged adenoids (Israel, 1962, Sato et al., 2009, Herberhold et al.,

2009) and emphasised the challenge of identification and recruitment of suitable

control patients. The specific roles of adenoidal viral flora and concurrent bacterial

infection in OM pathogenesis, through stimulation of host immune responses

remain a key direction for ongoing research.

Host immune responses have a central role OM pathogenesis and there is

controversy as to whether children with RAOM/COME have compromised or

dysfunctional immune responses. Mounting a successful immune response to

conserved antigens from the otopathogenic bacteria is essential for development

of effective vaccine-based prevention strategies. This study showed that there

239

were no significant differences in the levels of systemic and local antibodies

developed against protein vaccine candidates of S. pneumoniae and H.

influenzae. Indeed, the level of anti-P4 IgA antibody in serum of children with

RAOM/COME was significantly higher than that observed in children without

RAOM/COME. Given the controversy identified in previous reports, the current

results contradict previous studies which demonstrated that children with more

episodes of OM mount a less successful host immune response, resulting in

reduced levels of IgG and IgA against antigens of S. pneumoniae (Freijd et al.,

1984, Veenhoven et al., 2004, Prellner et al., 1989, Dhooge et al., 2002, Kaur et

al., 2011a) and of H. influenzae (Pichichero et al., 2010, Kaur et al., 2011b),

compared to children with less episodes of OM. In contrast, the current results

provide additional support to previous studies that reported no evidence of

impaired serum antibody responses to antigens of S. pneumoniae and H.

influenzae from children the with RAOM (Berman et al., 1992, Drake-Lee et al.,

2003, Misbah et al., 1997, Sørensen and Nielsen, 1988, Wiertsema et al., 2012,

Menon et al., 2012, Corscadden et al., 2013).

We further explored the potential mechanisms of the host immune responses

mounted using systemic gene expression analysis to compare RAOM/COME

prone and control groups of children and showed no significant differences in the

immuno-gene expression between the groups (Chapter 6). Interestingly, IgA and

IgG antibodies against antigens of S. pneumoniae and H. influenzae were also

detected frequently in middle ear of children with RAOM/COME, which is

consistent with a previous study (Corscadden et al., 2013). This result provides

further evidence that children with RAOM/COME mount normal antibody immune

responses. In combination with the frequency of bacterial detection identified in

the current study, the bactericidal capacity of these antibodies in middle ear were

suggested by the higher levels of antibodies detected within the MEE of children

with MEE negative for bacteria, compared to those observed in children with MEE

positive for bacteria. This result supports evidence for children experiencing

AOM, where the presence of antibodies within the middle ear was associated

with the clearance of bacteria (Howie et al., 1973, Sloyer et al., 1976, Karjalainen

et al., 1991b, Karjalainen et al., 1991a). Collectively, children with RAOM/COME

show a normal capacity to mount immune responses against infection. Therefore

240

development of a multi-pathogen microbial vaccine could reduce or prevent

microbial colonisation/infection of the URT and reduce the incidence of OM in at-

risk children.

Adaptive immune responses actioned through antibody production are integrated

and reliant upon initiation of innate immune responses induced through cytokine

cascades. In this study, the absence of evidence of cytokine responses detected

within the serum, confirmed that all children in the study were healthy at the time

of recruitment and sample collection. Detection of locally released cytokines

within the MEE of children with RAOM/COME demonstrated a high frequency of

pro-inflammatory (IL-1β, IL-6, IL-8 and TNF-α) and anti- inflammatory (IL-10)

cytokine detection which is consistent with other studies (Skotnicka and

Hassmann, 2000, Skotnicka and Hassmann, 2008, Zielnik-Jurkiewicz and

Stankiewicz-Szymczak, 2015) and evidences the activation of local innate

immune responses. Furthermore, the levels of cytokines measured in MEE

positive for microbes were significantly higher than what was observed in MEE

that were negative for microbes. These observations suggest that innate immune

responses were regulated locally in the middle ear in response to microbial

infection. Indeed, IL-10, is an important cytokine in immuno-regulation, that

inhibits Th1 function and promotes the activity of natural killer cells (Cai et al.,

1999, Conti et al., 2003, Pestka et al., 2004) and was strongly correlated with IL-

1β and TNF-α in the current study. Further evidence of the activation of the innate

immune response through cytokine cascades was provided through the systemic

gene expression analysis undertaken in this study. These analyses showed

higher levels of expression of genes related to innate immune responses in MEE

that were positive for microbes, compared to MEE negative for microbes

(Chapter 6). These observations provide further evidence of normal immune

responses being mounted by children with RAOM/COME.

Specific investigation of the potential genes impacting OM-prone children was

undertaken using systemic immuno-gene expression analysis. This study

identified that expression of HLA-DQA1 and HLA-DQB1 genes were differentially

expressed in children with and without RAOM/COME. This unique finding has no

precedent within the OM research literature. However HLA-DQA1 and HLA-

241

DQB1 genes have been associated with celiac disease, a congenital chronic

inflammatory condition of the small intestine (Dubois et al., 2010) and their role

in pathogenesis of this condition explored (Neuhausen et al., 2002, Murray et al.,

2007, Megiorni and Pizzuti, 2012). Although HLA-DQA1 and HLA-DQB1 genes

have not previously been identified or associated with OM, a gene that is closely

related to HLA-DQA1 and HLA-DQB1 genes, HLA-A2, has been associated with

RAOM in children (Kalm et al., 1991, Kalm et al., 1994). The potential role of

these genes in the pathogenesis of OM requires further investigation, including

assessment of gene expression levels and allelic variances with larger

populations to investigate their roles in chronicity and recurrence of OM.

To conclude, this study provides comprehensive data on the predominant

bacteria and viruses, local and systemic immune responses, and systemic

immuno-gene expression in children with and without RAOM/COME from urban

environments in Eastern Australia. H. influenzae and rhinovirus are the

predominant bacteria and virus detected in children with RAOM/COME,

respectively. The increase in the observed frequency of H. influenzae which

produce β-lactamase (December 2008-October 2015) needs to be considered

when prescribing β-lactam antibiotics, as failure to include inhibitors for this

enzyme may lead to treatment failures. Overall the prevalence of S. pneumoniae

in the URT was reduced and M. catarrhalis, increased respectively, in children

after PCV13 introduction and most S. pneumoniae were non-PCV13 serotypes.

Importantly, this study provided substantive evidence that normal immune

responses were observed in children with RAOM/COME and no evidence of

dysfunctional responses to the pathogens were evident. The identification of two

HLA-DQA1 and HLA-DQB1 genes that may play a role in development of

chronicity or recurrence of OM provides future directions for potentially identifying

children at increased risk of developing chronic or recurrent OM from an early

age. Together, these findings suggest that protein-based vaccines for S.

pneumoniae, H. influenzae and M. catarrhalis may benefit children by improved

prevention of URT colonisation/infection by these bacteria and subsequent

reduction in the incidence of OM. Inclusion of multiple predominant pathogens

within vaccine formulations for OM prevention is a key future direction for

protection of children at risk of chronic and/or recurrent otitis media.

242

Bibliography

ACCESS ECONOMICS 2009. The cost burden of otitis media in Australia. GlaxoSmithKline ACUIN, J. 2007. Chronic suppurative otitis media. BMJ Clin Evid., 2, 1-20. AHMED, S., SHAPIRO, N. & BHATTACHARYYA, N. 2014. Incremental health care utilization and

costs for acute otitis media in children. Laryngoscope., 124, 301-305. AL-MAZROU, K. & AL-KHATTAF, A. 2008. Adherent biofilms in adenotonsillar diseases in children.

Arch Otolaryngol Head Neck Surg., 134, 20-23. ALLEN, E., CHEN, W., WEEKS, D., CHEN, F., HOU, X., MATTOS, J., MYCHALECKYJ, J., SEGADE, F.,

CASSELBRANT, M., MANDEL, E., FERRELL, R., RICH, S., DALY, K. & SALE, M. 2013. A genome-wide association study of chronic otitis media with effusion and recurrent otitis media identifies a novel susceptibility locus on chromosome 2. J Assoc Res Otolaryngol., 14, 791-800.

ALLES, R., PARIKH, A., HAWK, L., DARBY, Y., ROMERO, J. & SCADDING, G. 2001. The prevalence of atopic disorders in children with chronic otitis media with effusion. Pediatr Allergy Immunol. , 12, 102-106.

ALMAC, A., ELICORA, S., YUMUK, Z., DUNDAR, V. & WILLKE, A. 2009. The relationship between chronic otitis media with effusion and surface and deep flora of hypertrophic adenoids. Int J Pediatr Otorhinolaryngol. , 73, 1438-1440.

ALONSO, M., MARIMON, J. M., ERCIBENGOA, M., PÉREZ-YARZA, E. G. & PÉREZ-TRALLERO, E. 2013. Dynamics of Streptococcus pneumoniae serotypes causing acute otitis media isolated from children with spontaneous middle-ear drainage over a 12-year period (1999-2010) in a region of northern Spain. PLoS One., 8, e54333.

ALPER, C., WINTHER, B., MANDEL, E., HENDLEY, J. & DOYLE, W. 2009. Rate of concurrent otitis media in upper respiratory tract infections with specific viruses. Arch Otolaryngol Head Neck Surg., 135, 17-21.

AMERICAN ACADEMY OF FAMILY PHYSICIANS 2004. Otitis media with effusion. Pediatrics, 113, 1412-1429.

ANGOULVANT, F., COHEN, R., DOIT, C., ELBEZ, A., WERNER, A., BÉCHET, S., BONACORSI, S., VARON, E. & LEVY, C. 2015. Trends in antibiotic resistance of Streptococcus pneumoniae and Haemophilus influenzae isolated from nasopharyngeal flora in children with acute otitis media in France before and after 13 valent pneumococcal conjugate vaccine introduction. BMC Infect Dis., 15.

ATKINSON, H., WALLIS, S. & COATESWORTH, A. 2015. Acute otitis media. Postgrad Med., 127, 386-390.

BAKALETZ, L., DANIELS, R. & LIM, D. 1993. Modeling adenovirus type 1-induced otitis media in the chinchilla: effect on ciliary activity and fluid transport function of eustachian tube mucosal epithelium. J Infect Dis., 168, 865-872.

BALDER, R., SHAFFER, T. & LAFONTAINE, E. 2013. Moraxella catarrhalis uses a twin-arginine translocation system to secrete the β-lactamase BRO-2. BMC Microbiol. , 13.

BANCHEREAU, J. & STEINMAN, R. 1998. Dendritic cells and the control of immunity. Nature., 392, 245-252.

BARKAI, G., LEIBOVITZ, E., GIVON-LAVI, N. & DAGAN, R. 2009. Potential contribution by nontypable Haemophilus influenzae in protracted and recurrent acute otitis media. Pediatr Infect Dis J., 28, 466-471.

BD PHARMINGEN. 2015. Technical Data Sheet: Transcription Factor Buffer Set. Available: http://www.bdbiosciences.com/ds/pm/tds/562725.pdf.

BEDER, L., HOTOMI, M., OGAMI, M., YAMAUCHI, K., SHIMADA, J., BILLAL, D., ISHIGURO, N. & YAMANAKA, N. 2009. Clinical and microbiological impact of human bocavirus on children with acute otitis media. Eur J Pediatr., 168, 1365-1372.

http://www.bdbiosciences.com/ds/pm/tds/562725.pdf

243

BEN-SHIMOL, S., GIVON-LAVI, N., GREENBERG, D. & DAGAN, R. 2016. Pneumococcal nasopharyngeal carriage in children <5 years of age visiting the pediatric emergency room in relation to PCV7 and PCV13 introduction in southern Israel. Hum Vaccin Immunother., 12, 268-276.

BEN-SHIMOL, S., GIVON-LAVI, N., LEIBOVITZ, E., RAIZ, S., GREENBERG, D. & DAGAN, R. 2014. Near-elimination of otitis media caused by 13-valent pneumococcal conjugate vaccine (PCV) serotypes in southern Israel shortly after sequential introduction of 7-valent/13-valent PCV. Clin Infect Dis., 59, 1724-1732.

BERMAN, S., LEE, B., NUSS, R., ROARK, R. & GICLAS, P. 1992. Immunoglobulin G, total and subclass, in children with or without recurrent otitis media. J Pediatr. , 121, 249-251.

BIALASIEWICZ, S., WHILEY, D., BUHRER-SKINNER, M., BAUTISTA, C., BARKER, K., AITKEN, S., GORDON, R., MULLER, R., LAMBERT, S., DEBATTISTA, J., NISSEN, M. & SLOOTS, T. 2009. A novel gel-based method for self-collection and ambient temperature postal transport of urine for PCR detection of Chlamydia trachomatis. Sex Transm Infect. , 85, 102-105.

BIESBROEK, G., WANG, X., KEIJSER, B., EIJKEMANS, R., TRZCIŃSKI, K., ROTS, N., VEENHOVEN, R., SANDERS, E. & BOGAERT, D. 2014. Seven-valent pneumococcal conjugate vaccine and nasopharyngeal microbiota in healthy children. Emerg Infect Dis., 20, 201-210.

BIILL PRIMO, O., LOURENÇO, E. & PASSOS, S. 2014. Detection of respiratory viruses in nasopharyngeal swab and adenoid tissue from children submitted to adenoidectomy: pre- and postoperative analysis. Int Arch Otorhinolaryngol., 18, 150-154.

BINKS, M., CHENG, A., SMITH-VAUGHAN, H., SLOOTS, T., NISSEN, M., WHILEY, D., MCDONNELL, J. & LEACH, A. 2011. Viral-bacterial co-infection in Australian Indigenous children with acute otitis media. BMC Infect Dis., 11.

BINKS, M., TEMPLE, B., KIRKHAM, L., WIERTSEMA, S., DUNNE, E., RICHMOND, P., MARSH, R., LEACH, A. & SMITH-VAUGHAN, H. 2012. Molecular surveillance of true nontypeable Haemophilus influenzae: an evaluation of PCR screening assays. PLoS One., 7, e34083.

BLACK, S., SHINEFIELD, H., FIREMAN, B., LEWIS, E., RAY, P., HANSEN, J., ELVIN, L., ENSOR, K., HACKELL, J., SIBER, G., MALINOSKI, F., MADORE, D., CHANG, I., KOHBERGER, R., WATSON, W., AUSTRIAN, R. & EDWARDS, K. 2000. Efficacy, safety and immunogenicity of heptavalent pneumococcal conjugate vaccine in children. Northern California Kaiser Permanente Vaccine Study Center Group. Pediatr Infect Dis J., 19, 187-195.

BLOCK, S., HEIKKINEN, T., TOBACK, S., ZHENG, W. & AMBROSE, C. 2011. The efficacy of live attenuated influenza vaccine against influenza-associated acute otitis media in children. Pediatr Infect Dis J., 30, 203-207.

BLUESTONE, C. 1996. Pathogenesis of otitis media: role of eustachian tube. Pediatr Infect Dis J., 15, 281-291.

BLUESTONE, C. 2003. Definitions, Terminology and Classification. In: ROSENFELD, R. & BLUESTONE, C. (eds.) Evidence-Based Otitis Media. Second ed. Hamilton, London: BC Decker Inc.

BOSWELL, J. & NIENHUYS, T. 1995. Onset of otitis media in the first eight weeks of life in aboriginal and non-aboriginal Australian infants. Ann Otol Rhinol Laryngol. , 104, 542-549.

BOSWELL, J. & NIENHUYS, T. 1996. Patterns of persistent otitis media in the first year of life in aboriginal and non-aboriginal infants. Ann Otol Rhinol Laryngol., 105, 893-900.

BRENNAN-JONES, C., WHITEHOUSE, A., PARK, J., HEGARTY, M., JACQUES, A., EIKELBOOM, R., SWANEPOEL, W., WHITE, J. & JAMIESON, S. 2015. Prevalence and risk factors for parent-reported recurrent otitis media during early childhood in the Western Australian Pregnancy Cohort (Raine) Study. J Paediatr Child Health., 51, 403-409.

BROIDES, A., DAGAN, R., GREENBERG, D., GIVON-LAVI, N. & LEIBOVITZ, E. 2009. Acute otitis media caused by Moraxella catarrhalis: epidemiologic and clinical characteristics. Clin Infect Dis. , 49, 1641-1647.

244

BUCKNER, J. 2010. Mechanisms of impaired regulation by CD4(+)CD25(+)FOXP3(+) regulatory T cells in human autoimmune diseases. Nat Rev Immunol., 10, 849-859.

BULUT, Y., GÜVEN, M., OTLU, B., YENIŞEHIRLI, G., ALADAĞ, I., EYIBILEN, A. & DOĞRU, S. 2007. Acute otitis media and respiratory viruses. Eur J Pediatr., 166, 223-228.

CAI, G., KASTELEIN, R. & HUNTER, C. 1999. IL-10 enhances NK cell proliferation, cytotoxicity and production of IFN-gamma when combined with IL-18. Eur J Immunol., 29, 2658-2665.

CAMPBELL, H., BONT, L. & NAIR, H. 2015. Respiratory syncytial virus (RSV) disease - new data needed to guide future policy. J Glob Health., 5, 020101.

CASEY, J., KAUER, R. & PICHICHERO, M. Otopathogens causing acute otitis media in the 13-Valent pneumococcal conjugate vaccine era. 18th International Symposium on Recent Advances in Otitis Media, 2015a National Harbor, MD, USA.

CASEY, J., KAUER, R. & PICHICHERO, M. 2015b. Otopathogens causing acute otitis media in the 13-valent pneumococcal conjugate vaccine era. in: 18th international symposium on recent advances in otitis media.

CASEY, J. R., ADLOWITZ, D. G. & PICHICHERO, M. E. 2010. New patterns in the otopathogens causing acute otitis media six to eight years after introduction of pneumococcal conjugate vaccine. The Pediatric Infectious Disease Journal, 29, 304-309.

CESANO, A. 2015. nCounter(®) PanCancer Immune Profiling Panel (NanoString Technologies, Inc., Seattle, WA). J Immunother Cancer., 3.

CHANTZI, F., PAPADOPOULOS, N., BAIRAMIS, T., TSIAKOU, M., BOURNOUSOUZIS, N., CONSTANTOPOULOS, A., LIAPI, G., XATZIPSALTI, M. & KAFETZIS, D. 2006. Human rhinoviruses in otitis media with effusion. Pediatr Allergy Immunol. , 17, 514-518.

CHEN, A., LI, H., HEBDA, P., ZEEVI, A. & SWARTS, J. 2005. Gene expression profiles of early pneumococcal otitis media in the rat. Int J Pediatr Otorhinolaryngol., 69, 1383-1393.

CHENG, J. & JAVIA, L. 2012. Methicillin-resistant Staphylococcus aureus (MRSA) pediatric tympanostomy tube otorrhea. Int J Pediatr Otorhinolaryngol. , 76, 1795-1798.

CHKHAIDZE, I., NEMSADZE, K., GOTSADZE, K., NIKOLEISHVILI, E. & GORDELADZE, G. 2007. Systemic inflammatory responses in patients with acute otitis media and the impact of treatment with sinupret. Georgian Med News. , 151, 40-44.

CHONMAITREE, T. 2000. Viral and bacterial interaction in acute otitis media. Pediatr Infect Dis J. , 19, S24-S30.

CHONMAITREE, T. 2006. Acute otitis media is not a pure bacterial disease. Clin Infect Dis. , 43, 1423-1425.

CHONMAITREE, T., ALVAREZ-FERNANDEZ, P., JENNINGS, K., TRUJILLO, R., MAROM, T., LOEFFELHOLZ, M., MILLER, A., MCCORMICK, D., PATEL, J. & PYLES, R. 2015. Symptomatic and asymptomatic respiratory viral infections in the first year of life: association with acute otitis media development. Clin Infect Dis. , 60, 1-9.

CHONMAITREE, T., REVAI, K., GRADY, J., CLOS, A., PATEL, J., NAIR, S., FAN, J. & HENRICKSON, K. 2008a. Viral upper respiratory tract infection and otitis media complication in young children. Clin Infect Dis, 46, 815-823.

CHONMAITREE, T., REVAI, K., GRADY, J., CLOS, A., PATEL, J., NAIR, S., FAN, J. & HENRICKSON, K. 2008b. Viral upper respiratory tract infection and otitis media complication in young children. Clin Infect Dis., 46, 815-823.

CHONMAITREE, T., RUOHOLA, A. & HENDLEY, J. 2012. Presence of viral nucleic acids in the middle ear: acute otitis media pathogen or bystander? Pediatr Infect Dis J., 31, 325-330.

CLEMENTS, D., LANGDON, L., BLAND, C. & WALTER, E. 1995. Influenza A vaccine decreases the incidence of otitis media in 6- to 30-month-old children in day care. Arch Pediatr Adolesc Med., 149, 1113-1117.

COATES, H., THORNTON, R., LANGLANDS, J., FILION, P., KEIL, A., VIJAYASEKARAN, S. & P, R. 2008. The role of chronic infection in children with otitis media with effusion: evidence for intracellular persistence of bacteria. Otolaryngol Head Neck Surg., 138, 778-781.

245

COHEN, R., VARON, E., DOIT, C., SCHLEMMER, C., ROMAIN, O., THOLLOT, F., BÉCHET, S., BONACORSI, S. & LEVY, C. 2015. A 13-year survey of pneumococcal nasopharyngeal carriage in children with acute otitis media following PCV7 and PCV13 implementation. Vaccine., 33, 5118-5126.

COLLINS, D., HOSKINS, A., BOWMAN, J., JONES, J., STEMBERGER, N., RICHMOND, P., LEACH, A. & D, L. 2013. High nasopharyngeal carriage of non-vaccine serotypes in Western Australian aboriginal people following 10 years of pneumococcal conjugate vaccination. PloS One, 8, e82280.

COMAR, M., ZANOTTA, N., ROSSI, T., PELOS, G. & D'AGARO, P. 2011. Secondary lymphoid tissue as an important site for WU polyomavirus infection in immunocompetent children. J Med Virol., 83, 1446-1450.

CONTI, P., KEMPURAJ, D., KANDERE, K., DI GIOACCHINO, M., BARBACANE, R., CASTELLANI, M., FELACO, M., BOUCHER, W., LETOURNEAU, R. & THEOHARIDES, T. 2003. IL-10, an inflammatory/inhibitory cytokine, but not always. Immunol Lett. , 86, 123-129.

CORSCADDEN, K., KIRKHAM, L., THORNTON, R., VIJAYASEKARAN, S., COATES, H., RICHMOND, P. & WIERTSEMA, S. 2013. High pneumococcal serotype specific IgG, IgG1 and IgG2 levels in serum and the middle ear of children with recurrent acute otitis media receiving ventilation tubes. Vaccine. , 31, 1393-1399.

COSTANTINI, A., MANCINI, S., GIULIODORO, S., BUTINI, L., REGNERY, C., SILVESTRI, G. & MONTRONI, M. 2003. Effects of cryopreservation on lymphocyte immunophenotype and function. J Immunol Methods., 278, 145-155.

COSTERTON, W., VEEH, R., SHIRTLIFF, M., PASMORE, M., POST, C. & EHRLICH, G. 2003. The application of biofilm science to the study and control of chronic bacterial infections. J Clin Invest., 112, 1466-1477.

COUPER, K., BLOUNT, D., WILSON, M., HAFALLA, J., BELKAID, Y., KAMANAKA, M., FLAVELL, R., DE SOUZA, J. & RILEY, E. 2008. IL-10 from CD4CD25Foxp3CD127 adaptive regulatory T cells modulates parasite clearance and pathology during malaria infection. PLoS Pathog., 4, e1000004.

COYTE, P., ASCHE, C. & ELDEN, L. 1999. The economic cost of otitis media in Canada. Int J Pediatr Otorhinolaryngol., 49, 27-36.

COYTE, P., CROXFORD, R., MCISAAC, W., FELDMAN, W. & FRIEDBERG, J. 2001. The role of adjuvant adenoidectomy and tonsillectomy in the outcome of the insertion of tympanostomy tubes. N Engl J Med. , 344, 1188-1195.

CRIPPS, A., OTCZYK, D. & KYD, J. 2005. Bacterial otitis media: a vaccine preventable disease? Vaccine., 23, 2304-2310.

CRIPPS, A. W. & KYD, J. 2003. Bacterial otitis media: current vaccine development strategies. Immunol Cell Biol, 81, 46-51.

CRIPPS, A. W. & OTCZYK, D. C. 2006. Prospects for a vaccine against otitis media. Expert Rev Vaccines, 5, 517-534.

DAGAN, R., PELTON, S., BAKALETZ, L. & COHEN, R. 2016. Prevention of early episodes of otitis media by pneumococcal vaccines might reduce progression to complex disease. Lancet Infect Dis., 16, 480-492.

DAMOISEAUX, R., VAN BALEN, F., HOES, A. & DE MELKER, R. 1998. Antibiotic treatment of acute otitis media in children under two years of age: evidence based? Br J Gen Pract. , 48, 1861-1864.

DANG, X., BIALASIEWICZ, S., NISSEN, M., SLOOTS, T., KORALNIK, I. & TAN, C. 2011. Infrequent detection of KI, WU and MC polyomaviruses in immunosuppressed individuals with or without progressive multifocal leukoencephalopathy. PLoS One. , 6, e16736.

DARE, R., FRY, A., CHITTAGANPITCH, M., SAWANPANYALERT, P., OLSEN, S. & ERDMAN, D. 2007. Human coronavirus infections in rural Thailand: a comprehensive study using real-time reverse-transcription polymerase chain reaction assays. J Infect Dis. , 196, 1321-1328.

246

DARRIEUX, M., GOULART, C., BRILES, D. & LEITE, L. 2015. Current status and perspectives on protein-based pneumococcal vaccines. Crit Rev Microbiol. , 41, 190-200.

DE CHIARA, M., HOOD, D., MUZZI, A., PICKARD, D., PERKINS, T., PIZZA, M., DOUGAN, G., RAPPUOLI, R., MOXON, E., SORIANI, M. & DONATI, C. 2014. Genome sequencing of disease and carriage isolates of nontypeable Haemophilus influenzae identifies discrete population structure. Proc Natl Acad Sci U S A. , 111, 5439-5444.

DEL MAR, C., GLASZIOU, P. & HAYEM, M. 1997. Are antibiotics indicated as initial treatment for children with acute otitis media? A meta-analysis. BMJ., 314, 1526-1529.

DELGOFFE, G., WOO, S., TURNIS, M., GRAVANO, D., GUY, C., OVERACRE, A., BETTINI, M., VOGEL, P., FINKELSTEIN, D., BONNEVIER, J., WORKMAN, C. & VIGNALI, D. 2013. Stability and function of regulatory T cells is maintained by a neuropilin-1-semaphorin-4a axis. Nature., 501, 252-256.

DESAI, A., SHARMA, D., CRISPELL, E., BAUGHMAN, W., THOMAS, S., TUNALI, A., SHERWOOD, L., ZMITROVICH, A., JERRIS, R., SATOLA, S., BEALL, B., MOORE, M., JAIN, S. & FARLEY, M. 2015. Decline in Pneumococcal Nasopharyngeal Carriage of Vaccine Serotypes After the Introduction of the 13-Valent Pneumococcal Conjugate Vaccine in Children in Atlanta, Georgia. Pediatr Infect Dis J., 34, 1168-1174.

DEVINE, V., JEFFERIES, J., CLARKE, S. & FAUST, S. 2015. Nasopharyngeal Bacterial Carriage in the Conjugate Vaccine Era with a Focus on Pneumococci. J Immunol Res., 2015.

DHOOGE, I., VAN KEMPEN, M., SANDERS, L. & RIJKERS, G. 2002. Deficient IgA and IgG2 anti-pneumococcal antibody levels and response to vaccination in otitis prone children. Int J Pediatr Otorhinolaryngol. , 64, 133-141.

DICKO, A., SANTARA, G., MAHAMAR, A., SIDIBE, Y., BARRY, A., DICKO, Y. & AL, E. 2013. Safety, reactogenicity and immunogenicity of a booster dose of the 10-valent pneumococcal non-typeable Haemophilus influenzae protein D conjugate vaccine (PHiD-CV) in Malian children. Hum Vaccin Immunother., 9, 382 - 388.

DRAKE-LEE, A., HUGHES, R. & DUNN, C. 2003. Serum IgA and IgG functional antibodies and their subclasses to Streptococcus pneumoniae capsular antigen found in two aged-matched cohorts of children with and without otitis media with effusion. Clin Otolaryngol Allied Sci., 28, 335-340.

DRAWZ, S. & BONOMO, R. 2010. Three decades of beta-lactamase inhibitors. Clin Microbiol Rev., 23, 160-201.

DUBOIS, P., TRYNKA, G., FRANKE, L., HUNT, K., ROMANOS, J., CURTOTTI, A., ZHERNAKOVA, A., HEAP, G., ADÁNY, R., AROMAA, A., BARDELLA, M., VAN DEN BERG, L., BOCKETT, N., DE LA CONCHA, E., DEMA, B., FEHRMANN, R., FERNÁNDEZ-ARQUERO, M., FIATAL, S., GRANDONE, E., GREEN, P., GROEN, H., GWILLIAM, R., HOUWEN, R., HUNT, S., KAUKINEN, K., KELLEHER, D., KORPONAY-SZABO, I., KURPPA, K., MACMATHUNA, P., MÄKI, M., MAZZILLI, M., MCCANN, O., MEARIN, M., MEIN, C., MIRZA, M., MISTRY, V., MORA, B., MORLEY, K., MULDER, C., MURRAY, J., NÚÑEZ, C., OOSTEROM, E., OPHOFF, R., POLANCO, I., PELTONEN, L., PLATTEEL, M., RYBAK, A., SALOMAA, V., SCHWEIZER, J., SPERANDEO, M., TACK, G., TURNER, G., VELDINK, J., VERBEEK, W., WEERSMA, R., WOLTERS, V., URCELAY, E., CUKROWSKA, B., GRECO, L., NEUHAUSEN, S., MCMANUS, R., BARISANI, D., DELOUKAS, P., BARRETT, J., SAAVALAINEN, P., WIJMENGA, C. & VAN HEEL, D. 2010. Multiple common variants for celiac disease influencing immune gene expression. Nat Genet. , 42, 295-302.

DUPONT, D., MAHJUOB-MESSAI, F., FRANCOIS, M., DOIT, C., MARIANI-KURKDJIAN, P., BIDET, P., BONACORSI, S., CAROL, A. & BINGEN, E. 2010. Evolving microbiology of complicated acute otitis media before and after introduction of the pneumococcal conjugate vaccine in France. Diagnostic Microbiology and Infectious Disease, 68, 89-92.

247

EHRLICH, G., VEEH, R., WANG, X., COSTERTON, J., HAYES, J., HU, F., DAIGLE, B., EHRLICH, M. & POST, J. 2002. Mucosal biofilm formation on middle-ear mucosa in the chinchilla model of otitis media. JAMA, 287, 1710-1715.

ELKORD, E. 2009. Frequency of human T regulatory cells in peripheral blood is significantly reduced by cryopreservation. J Immunol Methods., 347, 87-90.

EMANEINI, M., GHARIBPOUR, F., KHORAMROOZ, S., MIRSALEHIAN, A., JABALAMELI, F., DARBAN-SAROKHALIL, D., MIRZAII, M., SHARIFI, A. & TAHERIKALANI, M. 2013. Genetic similarity between adenoid tissue and middle ear fluid isolates of Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis from Iranian children with otitis media with effusion. Int J Pediatr Otorhinolaryngol., 77, 1841-1845.

EMONTS, M., VEENHOVEN, R., WIERTSEMA, S., HOUWING-DUISTERMAAT, J., WALRAVEN, V., DE GROOT, R., HERMANS, P. & SANDERS, E. 2007. Genetic polymorphisms in immunoresponse genes TNFA, IL6, IL10, and TLR4 are associated with recurrent acute otitis media. Pediatrics., 120, 814-823.

ENGEL, J., ANTEUNIS, L., VOLOVICS, A., HENDRIKS, J. & MARRES, E. 1999. Risk factors of otitis media with effusion during infancy. Int J Pediatr Otorhinolaryngol. , 48, 239-249.

ESKOLA, J., KILPI, T., PALMU, A., JOKINEN, J., HAAPAKOSKI, J., HERVA, E., TAKALA, A., KÄYHTY, H., KARMA, P., KOHBERGER, R., SIBER, G., MÄKELÄ, P. & FINNISH OTITIS MEDIA STUDY GROUP 2001. Efficacy of a pneumococcal conjugate vaccine against acute otitis media. N Engl J Med., 344, 403-409.

EZAKI, T. & SUZUKI, S. 1982. Achromopeptidase for lysis of anaerobic gram-positive cocci. J Clin Microbiol. , 16, 844-846.

FADEN, H., DUFFY, L., WASIELEWSKI, R., WOLF, J., KRYSTOFIK, D. & TUNG, Y. 1997. Relationship between nasopharyngeal colonization and the development of otitis media in children. Tonawanda/Williamsville Pediatrics. J Infect Dis. , 175, 1440-1445.

FAINSTEIN, V., MUSHER, D. & CATE, T. 1980. Bacterial adherence to pharyngeal cells during viral infection. J Infect Dis., 141, 172-176.

FENOLL, A., AGUILAR, L., VICIOSO, M.-D., ROBLEDO, O. & GRANIZO, J.-J. 2011. Increase in serotype 19A prevalence and amoxicillin non-susceptibility among paediatric Streptococcus pneumoniae isolates from middle ear fluid in a passive laboratory-based surveillance in Spain, 1997-2009. BMC Infectious Diseases, 11, 1-8.

FREIJD, A., HAMMARSTRÖM, L., PERSSON, M. & SMITH, C. 1984. Plasma anti-pneumococcal antibody activity of the IgG class and subclasses in otitis prone children. Clin Exp Immunol., 56, 233-238.

GATES, G. 1999. Otitis media--the pharyngeal connection. JAMA., 282, 987-989. GEISS, G., BUMGARNER, R., BIRDITT, B., DAHL, T., DOWIDAR, N., DUNAWAY, D., FELL, H., FERREE,

S., GEORGE, R., GROGAN, T., JAMES, J., MAYSURIA, M., MITTON, J., OLIVERI, P., OSBORN, J., PENG, T., RATCLIFFE, A., WEBSTER, P., DAVIDSON, E., HOOD, L. & DIMITROV, K. 2008. Direct multiplexed measurement of gene expression with color-coded probe pairs. Nat Biotechnol. , 26, 317-325.

GENE, A., GARCIA-GARCIA, J. J., DOMINGO, A. & PALACIN, P. W. Y. E. 2004. Etiology of acute otitis media in a children's hospital and antibiotic sensitivity of the bacteria involved. Enferm Infecc Microbiol Clin, 22, 377-380.

GIEBINK, G., BERZINS, I., MARKER, S. & SCHIFFMAN, G. 1980. Experimental otitis media after nasal inoculation of Streptococcus pneumoniae and influenza A virus in chinchillas. Infect Immun., 30, 445-450.

GISSELSSON-SOLÉN, M., HENRIKSSON, G., HERMANSSON, A. & MELHUS, A. 2014. Risk factors for carriage of AOM pathogens during the first 3 years of life in children with early onset of acute otitis media. Acta Otolaryngol., 134, 684-690.

GLANVILLE, N. & JOHNSTON, S. 2015. Challenges in developing a cross-serotype rhinovirus vaccine. Curr Opin Virol. , 11, 83-88.

248

GRIJALVA, C., NUORTI, J. & GRIFFIN, M. 2009. Antibiotic prescription rates for acute respiratory tract infections in US ambulatory settings. JAMA., 302, 758-766.

GRUBB, M. S. & SPAUGH, D. C. 2010. Microbiology of acute otitis media, Puget Sound region, 2005-2009. Clinical Pediatrics, 49, 727-730.

GUNASEKERA, H., KNOX, S., MORRIS, P., BRITT, H., MCINTYRE, P. & CRAIG, J. 2007a. The spectrum and management of otitis media in Australian indigenous and nonindigenous children: a national study. Pediatr Infect Dis J., 26, 689-692.

GUNASEKERA, H., KNOX, S., MORRIS, P., BRITT, H., MCINTYRE, P. & CRAIG, J. 2007b. The spectrum and management of otitis media in Australian indigenous and nonindigenous children: a national study. Pediatr Infect Dis J., 26, 689-692.

GUNASEKERA, H., O'CONNOR, T., VIJAYASEKARAN, S. & DEL MAR, C. 2009. Primary care management of otitis media among Australian children. Med J Aust., 191, S55-S59.

GÜNEL, C., KıRDAR, S., ÖMÜRLÜ, İ. & AĞDAŞ, F. 2015. Detection of the Epstein-Barr virus, Human Bocavirus and novel KI and KU polyomaviruses in adenotonsillar tissues. Int J Pediatr Otorhinolaryngol., 79, 423-427.

HABIB, M., PORTER, B. & SATZKE, C. 2014. Capsular serotyping of Streptococcus pneumoniae using the Quellung reaction. J Vis Exp., 84, e51208.

HAFRÉN, L., EINARSDOTTIR, E., KENTALA, E., HAMMARÉN-MALMI, S., BHUTTA, M., MACARTHUR, C., WILMOT, B., CASSELBRANT, M., CONLEY, Y., WEEKS, D., MANDEL, E., VAARALA, O., KALLIO, A., MELIN, M., NIEMINEN, J., LEINONEN, E., KERE, J. & MATTILA, P. 2015. Predisposition to Childhood Otitis Media and Genetic Polymorphisms within the Toll-Like Receptor 4 (TLR4) Locus. PLoS One., 10, e0132551.

HALL-STOODLEY, L., HU, F., GIESEKE, A., NISTICO, L., NGUYEN, D., HAYES, J., FORBES, M., GREENBERG, D., DICE, B., BURROWS, A., WACKYM, P., STOODLEY, P., POST, J., EHRLICH, G. & JE, K. 2006. Direct detection of bacterial biofilms on the middle-ear mucosa of children with chronic otitis media. JAMA., 296, 202-211.

HALL-STOODLEY, L. & STOODLEY, P. 2002. Developmental regulation of microbial biofilms. Curr Opin Biotechnol., 13, 228-233.

HARABUCHI, Y., FADEN, H., YAMANAKA, N., DUFFY, L., WOLF, J. & KRYSTOFIK, D. 1994. Nasopharyngeal colonization with nontypeable Haemophilus influenzae and recurrent otitis media. Tonawanda/Williamsville Pediatrics. J Infect Dis., 170, 862-866.

HARE, K., SMITH-VAUGHAN, H., BEISSBARTH, J. & LEACH, A. 2015. Haemophilus influenzae isolates survive for up to 20 years at -70 °C in skim milk tryptone glucose glycerol broth (STGGB) if thawing is avoided during re-culture. J Microbiol Methods., 119, 132-133.

HARE, K., SMITH-VAUGHAN, H., BINKS, M., PARK, I., NAHM, M. & LEACH, A. 2009. "Dodgy 6As": differentiating pneumococcal serotype 6C from 6A by use of the Quellung reaction. J Clin Microbiol., 47, 1981-1982.

HARIMAYA, A., FUJII, N. & HIMI, T. 2009. Preliminary study of proinflammatory cytokines and chemokines in the middle ear of acute otitis media due to Alloiococcus otitidis. Int J Pediatr Otorhinolaryngol., 73, 677-680.

HARMES, K., BLACKWOOD, R., BURROWS, H., COOKE, J., HARRISON, R. & PASSAMANI, P. 2013. Otitis media: diagnosis and treatment. Am Fam Physician., 88, 435-440.

HEIKKINEN, T. 2000. Role of viruses in the pathogenesis of acute otitis media. Pediatr Infect Dis J. , 19, S17-S22.

HEIKKINEN, T., BLOCK, S., TOBACK, S., WU, X. & AMBROSE, C. 2013. Effectiveness of intranasal live attenuated influenza vaccine against all-cause acute otitis media in children. Pediatr Infect Dis J. , 32, 669-674.

HEIKKINEN, T. & CHONMAITREE, T. 2000. Increasing importance of viruses in acute otitis media. Ann Med., 32, 157-163.

HEIKKINEN, T. & CHONMAITREE, T. 2003. Importance of respiratory viruses in acute otitis media. Clin Microbiol Rev. , 16, 230-241.

249

HEIKKINEN, T., RUUSKANEN, O., WARIS, M., ZIEGLER, T., AROLA, M. & HALONEN, P. 1991. Influenza vaccination in the prevention of acute otitis media in children. Am J Dis Child. , 145, 445-448.

HEIKKINEN, T., THINT, M. & CHONMAITREE, T. 1999. Prevalence of various respiratory viruses in the middle ear during acute otitis media. Engl J Med., 340, 260-264.

HENDOLIN, P., MARKKANEN, A., YLIKOSKI, J. & JJ, W. 1997. Use of multiplex PCR for simultaneous detection of four bacterial species in middle ear effusions. J Clin Microbiol. , 35, 2854-2858.

HERBERHOLD, S., EIS-HÜBINGER, A. & PANNING, M. 2009. Frequent detection of respiratory viruses by real-time PCR in adenoid samples from asymptomatic children. J Clin Microbiol. , 47, 2682-2683.

HERNANDEZ, M., LEICHTLE, A., PAK, K., WEBSTER, N., WASSERMAN, S. & RYAN, A. 2015. The transcriptome of a complete episode of acute otitis media. BMC Genomics., 16.

HOA, M., SYAMAL, M., SCHAEFFER, M., SACHDEVA, L., BERK, R. & COTICCHIA, J. 2010. Biofilms and chronic otitis media: an initial exploration into the role of biofilms in the pathogenesis of chronic otitis media. Am J Otolaryngol. , 31, 241-245.

HOA, M., TOMOVIC, S., NISTICO, L., HALL-STOODLEY, L., STOODLEY, P., SACHDEVA, L., BERK, R. & COTICCHIA, J. 2009. Identification of adenoid biofilms with middle ear pathogens in otitis-prone children utilizing SEM and FISH. Int J Pediatr Otorhinolaryngol. , 73, 1242-1248.

HOBERMAN, A., GREENBERG, D., PARADISE, J., ROCKETTE, H., LAVE, J., KEARNEY, D., COLBORN, D., KURS-LASKY, M., HARALAM, M., BYERS, C., ZOFFEL, L., FABIAN, I., BERNARD, B. & KERR, J. 2003. Effectiveness of inactivated influenza vaccine in preventing acute otitis media in young children: a randomized controlled trial. JAMA. , 290, 1608-1616.

HOBERMAN, A., PARADISE, J., SHAIKH, N., GREENBERG, D., KEARNEY, D., COLBORN, D., ROCKETTE, H., KURS-LASKY, M., MCELLISTREM, M., ZOFFEL, L., BALENTINE, T., BARBADORA, K. & WALD, E. 2011. Pneumococcal resistance and serotype 19A in Pittsburgh-area children with acute otitis media before and after introduction of 7-valent pneumococcal polysaccharide vaccine. Clin Pediatr (Phila). , 50, 114-120.

HOLDER, R., KIRSE, D., EVANS, A., WHIGHAM, A., PETERS, T., POEHLING, K., SWORDS, W. & REID, S. 2015. Otopathogens Detected in Middle Ear Fluid Obtained during Tympanostomy Tube Insertion: Contrasting Purulent and Non-Purulent Effusions. PLoS One. , 10, e0128606.

HONG, H., KIM, T. & CHUNG, J. 2014. Long-term follow-up of otitis media with effusion in children: comparisons between a ventilation tube group and a non-ventilation tube group. Int J Pediatr Otorhinolaryngol., 78, 938-943.

HOSKINS, J., ALBORN, W. J., ARNOLD, J., BLASZCZAK, L., BURGETT, S., DEHOFF, B., ESTREM, S., FRITZ, L., FU, D., FULLER, W., GERINGER, C., GILMOUR, R., GLASS, J., KHOJA, H., KRAFT, A., LAGACE, R., LEBLANC, D., LEE, L., LEFKOWITZ, E., LU, J., MATSUSHIMA, P., MCAHREN, S., MCHENNEY, M., MCLEASTER, K., MUNDY, C., NICAS, T., NORRIS, F., O'GARA, M., PEERY, R., ROBERTSON, G., ROCKEY, P., SUN, P., WINKLER, M., YANG, Y., YOUNG-BELLIDO, M., ZHAO, G., ZOOK, C., BALTZ, R., JASKUNAS, S., ROSTECK, P. J., SKATRUD, P. & GLASS, J. 2001. Genome of the bacterium Streptococcus pneumoniae strain R6. J Bacteriol. , 183, 5709-5717.

HOTOMI, M., SAMUKAWA, T. & YAMANAKA, N. 1994. Interleukin-8 in otitis media with effusion. Acta Otolaryngol. , 114, 406-409.

HOWIE, V., PLOUSSARD, J., SLOYER, J. & JOHNSTON, R. J. 1973. Immunoglobulins of the middle ear fluid in acute otitis media: relationship to serum immunoglobulin concentrations and bacterial cultures. Infect Immun. , 7, 589-593.

250

ILIA, S., GOULIELMOS, G., SAMONIS, G. & GALANAKIS, E. 2014. Polymorphisms in IL-6, IL-10, TNF-α, IFN-γ and TGF-β1 genes and susceptibility to acute otitis media in early infancy. Pediatr Infect Dis J. , 33, 518-521.

IMANISHI, J. 2000. Expression of cytokines in bacterial and viral infections and their biochemical aspects. J Biochem., 127, 525-530.

ISRAEL, M. 1962. The viral flora of enlarged tonsils and adenoids. The Journal of Pathology and Bacteriology, 84, 169-176.

JACOBY, P., CARVILLE, K., HALL, G., RILEY, T., BOWMAN, J., LEACH, A., LEHMANN, D. & TEAM, K. O. M. R. P. 2011. Crowding and other strong predictors of upper respiratory tract carriage of otitis media-related bacteria in Australian Aboriginal and non-Aboriginal children. Pediatr Infect Dis J. , 30, 840-485.

JACOBY, P., COATES, H., ARUMUGASWAMY, A., ELSBURY, D., STOKES, A., MONCK, R., FINUCANE, J., WEEKS, S. & D, L. 2008. The effect of passive smoking on the risk of otitis media in Aboriginal and non-Aboriginal children in the Kalgoorlie-Boulder region of Western Australia. Med J Aust., 188, 599-603.

JAUNEIKAITE, E., TOCHEVA, A., JEFFERIES, J., GLADSTONE, R., FAUST, S., CHRISTODOULIDES, M., HIBBERD, M. & CLARKE, S. 2015. Current methods for capsular typing of Streptococcus pneumoniae. J Microbiol Methods. , 113.

JEFFERIES, J. M. C., MACDONALD, E., FAUST, S. N. & CLARKE, S. C. 2011. 13-valent pneumococcal conjugate vaccine (PCV13). Human Vaccines, 7, 1012-1018.

JEFFERSON, T., SMITH, S., DEMICHELI, V., HARNDEN, A., RIVETTI, A. & DI PIETRANTONJ, C. 2005. Assessment of the efficacy and effectiveness of influenza vaccines in healthy children: systematic review. Lancet., 365, 773-780.

JERVIS-BARDY, J., ROGERS, G., MORRIS, P., SMITH-VAUGHAN, H., NOSWORTHY, E., LEONG, L., SMITH, R., WEYRICH, L., DE HAAN, J., CARNEY, A., LEACH, A., O'LEARY, S. & MARSH, R. 2015. The microbiome of otitis media with effusion in Indigenous Australian children. Int J Pediatr Otorhinolaryngol. , 79, 1548-1555.

JERVIS-BARDY, J., SANCHEZ, L. & CARNEY, A. 2014. Otitis media in Indigenous Australian children: review of epidemiology and risk factors. J Laryngol Otol., 128, S16-27.

JOHN, M., DUNNE, E., LICCIARDI, P., SATZKE, C., WIJBURG, O., ROBINS-BROWNE, R. & O'LEARY, S. 2013. Otitis media among high-risk populations: can probiotics inhibit Streptococcus pneumoniae colonisation and the risk of disease? Eur J Clin Microbiol Infect Dis., 32, 1101-1110.

JOHNSON, D., D'ONISE, K., HOLLAND, R., RAUPACH, J. & KOEHLER, A. 2012. Pneumococcal disease in South Australia: vaccine success but no time for complacency. Vaccine, 30, 2206-2211.

JOKINEN, J., RINTA-KOKKO, H., SIIRA, L., PALMU, A., VIRTANEN, M., NOHYNEK, H., VIROLAINEN-JULKUNEN, A., TOROPAINEN, M. & NUORTI, J. 2015. Impact of ten-valent pneumococcal conjugate vaccination on invasive pneumococcal disease in Finnish children--a population-based study. PLoS One., 10, e0120290.

JONES, E. J., THOMAS, L. & DAVIS, N. 1979. The significance of secretory IgA in middle ear fluid. Ann Allergy., 42, 236-240.

JUHN, S., GARVIS, W., LEES, C., LE, C. & KIM, C. 1994. Determining otitis media severity from middle ear fluid analysis. Ann Otol Rhinol Laryngol Suppl., 163, 43-45.

JUHN, S., TOLAN, C., GARVIS, W., CROSS, D. & GIEBINK, G. 1992. The levels of IL-1 beta in human middle ear effusions. Acta Otolaryngol Suppl., 493, 37-42.

KADHIM, A., SPILSBURY, K., SEMMENS, J., COATES, H. & LANNIGAN, F. 2007. Adenoidectomy for middle ear effusion: a study of 50,000 children over 24 years. Laryngoscope. , 117, 427-433.

251

KAIJALAINEN, T. & PALMU, A. 2015. Long-term survival of Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis as isolates and in nasopharyngeal specimens in frozen STGG storage medium. J Microbiol Methods., 114, 38-39.

KAIJALAINEN, T., RUOKOKOSKI, E., UKKONEN, P. & HERVA, E. 2004. Survival of Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis frozen in skim milk- tryptone-glucose-glycerol medium. J Clin Microbiol. , 42, 412-414.

KALIA, A., RATTAN, A. & CHOPRA, P. 1999. A method for extraction of high-quality and high-quantity genomic DNA generally applicable to pathogenic bacteria. Anal Biochem., 275, 1-5.

KALM, O., JOHNSON, U. & PRELLNER, K. 1994. HLA frequency in patients with chronic secretory otitis media. Int J Pediatr Otorhinolaryngol. , 30, 151-157.

KALM, O., JOHNSON, U., PRELLNER, K. & NINN, K. 1991. HLA frequency in patients with recurrent acute otitis media. Arch Otolaryngol Head Neck Surg., 117, 1296-1299.

KANIA, R., LAMERS, G., VONK, M., DORPMANS, E., STRUIK, J., TRAN BA HUY, P., HIEMSTRA, P., BLOEMBERG, G. & GROTE, J. 2008. Characterization of mucosal biofilms on human adenoid tissues. Laryngoscope. , 118, 128-134.

KAPLAN, S., BARSON, W., LIN, P., ROMERO, J., BRADLEY, J., TAN, T., HOFFMAN, J., GIVNER, L. & MASON, E. 2013. Early trends for invasive pneumococcal infections in children after the introduction of the 13-valent pneumococcal conjugate vaccine. Pediatr Infect Dis J., 32, 203-207.

KAPLAN, S., CENTER, K., BARSON, W., LING-LIN, P., ROMERO, J., BRADLEY, J., TAN, T., HOFFMAN, J., PETERS, T., GURTMAN, A., SCOTT, D., TRAMMEL, J., GRUBER, W., HULTEN, K. & MASON, E. 2015. Multicenter surveillance of Streptococcus pneumoniae isolates from middle ear and mastoid cultures in the 13-valent pneumococcal conjugate vaccine era. Clin Infect Dis., 60, 1339-1345.

KARJALAINEN, H., KOSKELA, M., LUOTONEN, J., HERVA, E. & SIPILÄ, P. 1991a. Occurrences of antibodies against Streptococcus pneumoniae, Haemophilus influenzae and Branhamella catarrhalis in middle ear effusion and serum during the course of acute otitis media. Acta Otolaryngol. , 111, 112-119.

KARJALAINEN, H., KOSKELA, M., LUOTONEN, J. & SIPILÄ, P. 1991b. Secretory antibodies specific to Streptococcus pneumoniae, Haemophilus influenzae and Branhamella catarrhalis in middle ear effusion during acute otitis media. Acta Otolaryngol., 111, 524-529.

KARLIDAĞ, T., DEMIRDAĞ, K., KAYGUSUZ, I., OZDEN, M., YALÇIN, S. & OZTÜRK, L. 2002. Resistant bacteria in the adenoid tissues of children with otitis media with effusion. Int J Pediatr Otorhinolaryngol., 64, 35-40.

KAUR, R., CASEY, J. & PICHICHERO, M. 2011a. Serum antibody response to five Streptococcus pneumoniae proteins during acute otitis media in otitis-prone and non-otitis-prone children. Pediatr Infect Dis J., 30, 645-650.

KAUR, R., CASEY, J. & PICHICHERO, M. 2011b. Serum antibody response to three non-typeable Haemophilus influenzae outer membrane proteins during acute otitis media and nasopharyngeal colonization in otitis prone and non-otitis prone children. Vaccine., 29, 1023-1028.

KAUR, R., CASEY, J. & PICHICHERO, M. 2013. Relationship with original pathogen in recurrence of acute otitis media after completion of amoxicillin/clavulanate: bacterial relapse or new pathogen. Pediatr Infect Dis J. , 32, 1159-1162.

KAWABE, T., MATSUSHIMA, M., HASHIMOTO, N., IMAIZUMI, K. & HASEGAWA, Y. 2011. CD40/CD40 ligand interactions in immune responses and pulmonary immunity. Nagoya J Med Sci. , 73, 69-78.

KERSCHNER, J., HORSEY, E., AHMED, A., ERBE, C., KHAMPANG, P., CIOFFI, J., HU, F., POST, J. & EHRLICH, G. 2009. Gene expression differences in infected and noninfected middle ear complementary DNA libraries. Arch Otolaryngol Head Neck Surg., 135, 33-39.

252

KEYNAN, Y., CARD, C., MCLAREN, P., DAWOOD, M., KASPER, K. & FOWKE, K. 2008. The role of regulatory T cells in chronic and acute viral infections. Clin Infect Dis., 46, 1046-1052.

KILPI, T., HERVA, E., KAIJALAINEN, T., SYRJANEN, R. & TAKALA, A. K. 2001. Bacteriology of acute otitis media in a cohort of Finnish children followed for the first two years of life. Pediatr Infect Dis J, 20, 654-662.

KIM, C. H., KIM, J. S., CHA, S. H., KIM, K. N., KIM, J. D., LEE, K. Y. & AL, E. 2011. Response to primary and booster vaccination with 10-valent pneumococcal nontypeable Haemophilus influenzae protein D conjugate vaccine in Korean infants. Pediatr Infect Dis J., 30, e235-e243.

KIRKHAM, L., WIERTSEMA, S., MOWE, E., BOWMAN, J., RILEY, T. & RICHMOND, P. 2010. Nasopharyngeal carriage of Haemophilus haemolyticus in otitis-prone and healthy children. J Clin Microbiol., 48, 2557-2559.

KLAASSEN, E., VAN DE KANT, K., JÖBSIS, Q., VAN SCHAYCK, O., SMOLINSKA, A., DALLINGA, J., VAN SCHOOTEN, F., DEN HARTOG, G., DE JONGSTE, J., RIJKERS, G. & DOMPELING, E. 2015. Exhaled biomarkers and gene expression at preschool age improve asthma prediction at 6 years of age. Am J Respir Crit Care Med. , 191, 201-207.

KLEIN, J. 2000. The burden of otitis media. Vaccine, 19, S2-S8. KONG, K. & COATES, H. L. C. 2009. Natural history, definitions, risk factors and burden of otitis

media. Med J Aust., 191, s39-s43. KOOPMAN, L., HOES, A., GLASZIOU, P., APPELMAN, C., BURKE, P., MCCORMICK, D.,

DAMOISEAUX, R., LE SAUX, N. & ROVERS, M. 2008. Antibiotic therapy to prevent the development of asymptomatic middle ear effusion in children with acute otitis media: a meta-analysis of individual patient data. Arch Otolaryngol Head Neck Surg. , 134, 128-132.

KORISTKA, S., CARTELLIERI, M., ARNDT, C., FELDMANN, A., TÖPFER, K., MICHALK, I., TEMME, A., EHNINGER, G. & BACHMANN, M. 2014. Cytotoxic response of human regulatory T cells upon T-cell receptor-mediated activation: a matter of purity. Blood Cancer J., 4.

KOTOWSKI, M., NIEDZIELSKI, A., NIEDZIELSKA, G. & LACHOWSKA-KOTOWSKA, P. 2011. Dendritic cells and lymphocyte subpopulations of the adenoid in the pathogenesis of otitis media with effusion. Int J Pediatr Otorhinolaryngol., 75, 265-269.

KRÄMER, A., GREEN, J., POLLARD, J. J. & TUGENDREICH, S. 2014. Causal analysis approaches in Ingenuity Pathway Analysis. Bioinformatics., 30, 523-530.

KREINER-MØLLER, E., CHAWES, B., CAYE-THOMASEN, P., BØNNELYKKE, K. & BISGAARD, H. 2012. Allergic rhinitis is associated with otitis media with effusion: a birth cohort study. Clin Exp Allergy., 42, 1615-1620.

KRISHNAMURTHY, A., MCGRATH, J., CRIPPS, A. & KYD, J. 2009. The incidence of Streptococcus pneumoniae otitis media is affected by the polymicrobial environment particularly Moraxella catarrhalis in a mouse nasal colonisation model. Microbes Infect., 11, 545-553.

KROGULSKA, A., BOROWIEC, M., POLAKOWSKA, E., DYNOWSKI, J., MŁYNARSKI, W. & WASOWSKA-KRÓLIKOWSKA, K. 2011. FOXP3, IL-10, and TGF-β genes expression in children with IgE-dependent food allergy. J Clin Immunol., 31, 205-215.

KUBBA, H., PEARSON, J. & BIRCHALL, J. 2000. The aetiology of otitis media with effusion: a review. Clin Otolaryngol Allied Sci., 25, 181-194.

LAGOS, R. E., MUÑOZ, A. E., LEVINE, M. M., LEPETIC, A., FRANÇOIS, N., YARZABAL, J. P. & SCHUERMAN, L. 2011. Safety and immunogenicity of the 10-valent pneumococcal nontypeable Haemophilus influenzae protein D conjugate vaccine (PHiD-CV) in Chilean children. Hum Vaccin., 7, 511-522.

LAMBERT, S., ALLEN, K., DRUCE, J., BIRCH, C., MACKAY, I., CARLIN, J., CARAPETIS, J., SLOOTS, T., NISSEN, M. & NOLAN, T. 2007. Community epidemiology of human metapneumovirus,

253

human coronavirus NL63, and other respiratory viruses in healthy preschool-aged children using parent-collected specimens. Pediatrics., 120, e929-e937.

LAMBERT, S., WARE, R., COOK, A., MAGUIRE, F., WHILEY, D., BIALASIEWICZ, S., MACKAY, I., WANG, D., SLOOTS, T., NISSEN, M. & GRIMWOOD, K. 2012. Observational Research in Childhood Infectious Diseases (ORChID): a dynamic birth cohort study. BMJ Open. , 2.

LEACH, A., BOSWELL, J., ASCHE, V., NIENHUYS, T. & MATHEWS, J. 1994. Bacterial colonization of the nasopharynx predicts very early onset and persistence of otitis media in Australian aboriginal infants. Pediatr Infect Dis J. , 13, 983-989.

LEACH, A., MORRIS, P., MATHEWS, J. & CHRONIC OTITIS MEDIA INTERVENTION TRIAL - ONE (COMIT1) GROUP 2008. Compared to placebo, long-term antibiotics resolve otitis media with effusion (OME) and prevent acute otitis media with perforation (AOMwiP) in a high-risk population: a randomized controlled trial. BMC Pediatr., 8.

LEACH, A., WIGGER, C., HARE, K., HAMPTON, V., BEISSBARTH, J., ANDREWS, R., CHATFIELD, M., SMITH-VAUGHAN, H. & MORRIS, P. 2015. Reduced middle ear infection with non-typeable Haemophilus influenzae, but not Streptococcus pneumoniae, after transition to 10-valent pneumococcal non-typeable H. influenzae protein D conjugate vaccine. BMC Pediatr. , 15.

LEHMANN, D., ARUMUGASWAMY, A., ELSBURY, D., FINUCANE, J., STOKES, A., MONCK, R., JEFFRIES-STOKES, C., MCAULLAY, D., COATES, H. & STANLEY, F. 2008. The Kalgoorlie Otitis Media Research Project: rationale, methods, population characteristics and ethical considerations. Paediatr Perinat Epidemiol. , 22, 60-71.

LEHTORANTA, L., SÖDERLUND-VENERMO, M., NOKSO-KOIVISTO, J., TOIVOLA, H., BLOMGREN, K., HATAKKA, K., POUSSA, T., KORPELA, R. & PITKÄRANTA, A. 2012. Human bocavirus in the nasopharynx of otitis-prone children. Int J Pediatr Otorhinolaryngol., 76, 206-211.

LEIBOVITZ, E., DAGAN, R., LAVER, J., PIGLANSKY, L., RAIZ, S., ABBOUD, M., FLISS, D., LEIBERMAN, A. & BARZILAI, A. 2000. Interleukin 8 in middle ear fluid during acute otitis media: correlation with aetiology and bacterial eradication. Arch Dis Child., 82, 165-168.

LEMON, K., KLEPAC-CERAJ, V., SCHIFFER, H., BRODIE, E., LYNCH, S. & KOLTER, R. 2010. Comparative analyses of the bacterial microbiota of the human nostril and oropharynx. MBio., 1, e00129-10.

LIEBERTHAL, A., CARROLL, A., CHONMAITREE, T., GANIATS, T., HOBERMAN, A., JACKSON, M., JOFFE, M., MILLER, D., ROSENFELD, R., SEVILLA, X., SCHWARTZ, R., THOMAS, P. & TUNKEL, D. 2013. The diagnosis and management of acute otitis media. Pediatrics., 131, e964-999.

LIU, K., CHEN, L., KAUR, R. & PICHICHERO, M. 2012. Transcriptome signature in young children with acute otitis media due to Streptococcus pneumoniae. Microbes Infect. , 14, 600-609.

LIU, K., CHEN, L., KAUR, R. & PICHICHERO, M. 2013a. Transcriptome signature in young children with acute otitis media due to non-typeable Haemophilus influenzae. Int Immunol. , 25, 353-361.

LIU, K., KAUR, R., ALMUDEVAR, A. & PICHICHERO, M. 2013b. Higher serum levels of interleukin 10 occur at onset of acute otitis media caused by Streptococcus pneumoniae compared to Haemophilus influenzae and Moraxella catarrhalis. Laryngoscope., 123, 1500-1505.

LOEFFELHOLZ, M., TRUJILLO, R., PYLES, R., MILLER, A., ALVAREZ-FERNANDEZ, P., PONG, D. & CHONMAITREE, T. 2014. Duration of rhinovirus shedding in the upper respiratory tract in the first year of life. Pediatrics. , 134, 1144-1150.

LOUS, J., BURTON, M., FELDING, J., OVESEN, T., ROVERS, M. & WILLIAMSON, I. 2005. Grommets (ventilation tubes) for hearing loss associated with otitis media with effusion in children. Cochrane Database Syst Rev., 1.

254

LOUS, J., RYBORG, C. & THOMSEN, J. 2011. A systematic review of the effect of tympanostomy tubes in children with recurrent acute otitis media. Int J Pediatr Otorhinolaryngol. , 75, 1058-1061.

LU, X., HOLLOWAY, B., DARE, R., KUYPERS, J., YAGI, S., WILLIAMS, J., HALL, C. & ERDMAN, D. 2008. Real-time reverse transcription-PCR assay for comprehensive detection of human rhinoviruses. J Clin Microbiol., 46, 533-539.

MACKENZIE, G., CARAPETIS, J., LEACH, A. & MORRIS, P. 2009. Pneumococcal vaccination and otitis media in Australian Aboriginal infants: comparison of two birth cohorts before and after introduction of vaccination. BMC Pediatr., 9.

MAKARI, D., STAAT, M., HENRICKSON, K., WU, X. & AMBROSE, C. 2015. The Underrecognized Burden of Respiratory Syncytial Virus Among Infants Presenting to US Emergency Departments. Clin Pediatr (Phila). 54, 594-597.

MALLONE, R., MANNERING, S., BROOKS-WORRELL, B., DURINOVIC-BELLÓ, I., CILIO, C., WONG, F., SCHLOOT, N., T-CELL WORKSHOP COMMITTEE & IMMUNOLOGY OF DIABETES SOCIETY 2011. Isolation and preservation of peripheral blood mononuclear cells for analysis of islet antigen-reactive T cell responses: position statement of the T-Cell Workshop Committee of the Immunology of Diabetes Society. Clin Exp Immunol. , 163, 33-49.

MANGODT, T., VAN HERCK, M., NULLENS, S., RAMET, J., DE DOOY, J., JORENS, P. & DE WINTER, B. 2015. The role of Th17 and Treg responses in the pathogenesis of RSV infection. Pediatr Res. , 78, 483-491.

MARCHISIO, P., CLAUT, L., ROGNONI, A., ESPOSITO, S., PASSALI, D., BELLUSSI, L., DRAGO, L., POZZI, G., MANNELLI, S., SCHITO, G. & PRINCIPI, N. 2003. Differences in nasopharyngeal bacterial flora in children with nonsevere recurrent acute otitis media and chronic otitis media with effusion: implications for management. Pediatr Infect Dis J., 22, 262-268.

MARCHISIO, P., ESPOSITO, S., BIANCHINI, S., DUSI, E., FUSI, M., NAZZARI, E., PICCHI, R., GALEONE, C. & PRINCIPI, N. 2009. Efficacy of injectable trivalent virosomal-adjuvanted inactivated influenza vaccine in preventing acute otitis media in children with recurrent complicated or noncomplicated acute otitis media. Pediatr Infect Dis J., 28, 855-859.

MAROM, T., NOKSO-KOIVISTO, J. & CHONMAITREE, T. 2012. Viral-bacterial interactions in acute otitis media. Curr Allergy Asthma Rep. , 12, 551-558.

MARSH, R. L., BINKS, M. J., BEISSBARTH, J., CHRISTENSEN, P., MORRIS, P. S., LEACH, A. J. & SMITH-VAUGHAN, H. C. 2012. Quantitative PCR of ear discharge from indigenous Australian children with acute otitis media with perforation supports a role for Alloiococcus otitidis as a secondary pathogen. BMC Ear, Nose and Throat Disorders, 12.

MARTIN, J., HOBERMAN, A., PARADISE, J., BARBADORA, K., SHAIKH, N., BHATNAGAR, S., SHOPE, T., BLOCK, S., HARALAM, M., KURS-LASKY, M., COLBORN, D. & GREEN, M. 2014a. Emergence of Streptococcus pneumoniae serogroups 15 and 35 in nasopharyngeal cultures from young children with acute otitis media. Pediatr Infect Dis J. , 33, e286-90.

MARTIN, J., HOBERMAN, A., PARADISE, J., BARBADORA, K., SHAIKH, N., BHATNAGAR, S., SHOPE, T., BLOCK, S., HARALAM, M., KURS-LASKY, M., COLBORN, D. & GREEN, M. 2014b. Emergence of Streptococcus pneumoniae serogroups 15 and 35 in nasopharyngeal cultures from young children with acute otitis media. Pediatr Infect Dis J., 33, e286-e290.

MARZOUK, H., NATHAWAD, R., HAMMERSCHLAG, M., WEEDON, J., BACHMAN, D. & GOLDSTEIN, N. 2011. Methicillin-resistant Staphylococcus aureus colonization in otitis-prone children. Arch Otolaryngol Head Neck Surg., 137, 1217-1222.

MASSA, H., LIM, D., KURONO, Y. & CRIPPS, A. 2015. Middle Ear and Eustachian Tube Mucosal Immunology. In: JIRI, M., WARREN, S., MICHAEL, W., HILDE, C., BART, N. & BRIAN, L. (eds.) Mucosal Immunology. 4th ed. Waltham, MA: Elsevier Inc.

MASSA, H. M., CRIPPS, A. W. & LEHMANN, D. 2009. Otitis media: viruses, bacteria, biofilms and vaccines. Medical journal of Australia, 191, s44-s49.

255

MATA, M., MILIAN, L., ARMENGOT, M. & CARDA, C. 2014. Gene mutations in primary ciliary dyskinesia related to otitis media. Curr Allergy Asthma Rep. , 14, 420.

MATSUZAKI, Y., TAKASHITA, E., OKAMOTO, M., MIZUTA, K., ITAGAKI, T., KATSUSHIMA, F., KATSUSHIMA, Y., NAGAI, Y. & NISHIMURA, H. 2009. Evaluation of a new rapid antigen test using immunochromatography for detection of human metapneumovirus in comparison with real-time PCR assay. J Clin Microbiol., 47, 2981-2984.

MATTILA, P., HAMMARÉN-MALMI, S., SAXEN, H., KAIJALAINEN, T., KÄYHTY, H. & TARKKANEN, J. 2010. Adenoidectomy and nasopharyngeal carriage of Streptococcus pneumoniae in young children. Arch Dis Child. , 95, 696-702.

MATTILA, P., HAMMARÉN-MALMI, S., SAXEN, H., KAIJALAINEN, T., KÄYHTY, H. & TARKKANEN, J. 2012. Adenoidectomy in young children and serum IgG antibodies to pneumococcal surface protein A and choline binding protein A. Int J Pediatr Otorhinolaryngol., 76, 1569-1574.

MAWAS, F., HO, M. M. & CORBEL, M. J. 2009. Current progress with Moraxella catarrhalis antigens as vaccine candidates. Expert Rev Vaccines., 8, 77-90.

MAXWELL, K., FITZGERALD, J., BURLESON, J., LEONARD, G., CARPENTER, R. & KREUTZER, D. 1994. Interleukin-8 expression in otitis media. Laryngoscope., 104, 989-995.

MAXWELL, K., LEONARD, G. & KREUTZER, D. 1997. Cytokine expression in otitis media with effusion. Tumor necrosis factor soluble receptor. Arch Otolaryngol Head Neck Surg., 123, 984-989.

MCCARTHY, M. & KOCKLER, D. 2004. Trivalent intranasal influenza vaccine, live. Ann Pharmacother., 38, 2086-2093.

MCCORMICK, D., CHANDLER, S. & CHONMAITREE, T. 2007. Laterality of acute otitis media: different clinical and microbiologic characteristics. Pediatr Infect Dis J., 26, 583-588.

MCCORMICK, D., GRADY, J., DIEGO, A., MATALON, R., REVAI, K., PATEL, J., HAN, Y. & CHONMAITREE, T. 2011. Acute otitis media severity: association with cytokine gene polymorphisms and other risk factors. Int J Pediatr Otorhinolaryngol., 75, 708-712.

MCERLEAN, P., SHACKELTON, L., ANDREWS, E., WEBSTER, D., LAMBERT, S., NISSEN, M., SLOOTS, T. & MACKAY, I. 2008. Distinguishing molecular features and clinical characteristics of a putative new rhinovirus species, human rhinovirus C (HRV C). PLoS One., 3, e1847.

MCGRATH, L., BECKER-DREPS, S., PATE, V. & BROOKHART, M. 2013. Trends in antibiotic treatment of acute otitis media and treatment failure in children, 2000-2011. PLoS One., 8, e81210.

MEGIORNI, F. & PIZZUTI, A. 2012. HLA-DQA1 and HLA-DQB1 in Celiac disease predisposition: practical implications of the HLA molecular typing. J Biomed Sci., 19.

MENON, V., CORSCADDEN, K., FUERY, A., THORNTON, R., KIRKHAM, L., RICHMOND, P. & WIERTSEMA, S. 2012. Children with otitis media mount a pneumococcal serotype specific serum IgG and IgA response comparable to healthy controls after pneumococcal conjugate vaccination. Vaccine. , 30, 3136-3144.

MILLS, K. 2004. Regulatory T cells: friend or foe in immunity to infection? Nat Rev Immunol. , 4, 841-855.

MILLS, N., BEST, E., MURDOCH, D., SOUTER, M., NEEFF, M., ANDERSON, T., SALKELD, L., AHMAD, Z., MAHADEVAN, M., BARBER, C., BROWN, C., WALKER, C. & WALLS, T. 2015. What is behind the ear drum? The microbiology of otitis media and the nasopharyngeal flora in children in the era of pneumococcal vaccination. J Paediatr Child Health., 51, 300-306.

MISBAH, S., GRIFFITHS, H., MITCHELL, T., FREELAND, A., HAENEY, M. & CHAPEL, H. 1997. Antipolysaccharide antibodies in 450 children with otitis media. Clin Exp Immunol., 109, 67-72.

MITTAL, R., ROBALINO, G., GERRING, R., CHAN, B., YAN, D., GRATI, M. & LIU, X. 2014. Immunity genes and susceptibility to otitis media: a comprehensive review. J Genet Genomics., 41, 567-581.

256

MIYAMOTO, N. & BAKALETZ, L. 1997. Kinetics of the ascension of NTHi from the nasopharynx to the middle ear coincident with adenovirus-induced compromise in the chinchilla. Microb Pathog., 23, 119-126.

MODRZYNSKI, M. & ZAWISZA, E. 2007. An analysis of the incidence of adenoid hypertrophy in allergic children. Int J Pediatr Otorhinolaryngol. , 71, 713-719.

MONASTA, L., RONFANI, L., MARCHETTI, F., MONTICO, M., BRUMATTI, L. V., BAVCAR, A., GRASSO, D., BARBIERO, C. & TAMBURLINI, G. 2012. Burden of Disease Caused by Otitis Media: Systematic Review and Global Estimates. PLos One, 7, 1-12.

MONOBE, H., ISHIBASHI, T., NOMURA, Y., SHINOGAMI, M. & YANO, J. 2003. Role of respiratory viruses in children with acute otitis media. Int J Pediatr Otorhinolaryngol. , 67, 801-806.

MOREIRA, M., CINTRA, O., HARRIAGUE, J., HAUSDORFF, W. & HOET, B. 2016. Impact of the introduction of the pneumococcal conjugate vaccine in the Brazilian routine childhood national immunization program. Vaccine. , 34, 2766-2778.

MORRIS, P., GADIL, G., MCCALLUM, G., WILSON, C., SMITH-VAUGHAN, H., TORZILLO, P. & LEACH, A. 2010. Single-dose azithromycin versus seven days of amoxycillin in the treatment of acute otitis media in Aboriginal children (AATAAC): a double blind, randomised controlled trial. Med J Aust., 192, 24-29.

MORRIS, P., LEACH, A., HALPIN, S., MELLON, G., GADIL, G., WIGGER, C., MACKENZIE, G., WILSON, C., GADIL, E. & TORZILLO, P. 2007. An overview of acute otitis media in Australian Aboriginal children living in remote communities. Vaccine., 25, 2389-2393.

MORRIS, P. S. & LEACH, A. J. 2009. acute and chronic otitis media. Pediatr Clin N Am, 56, 1383-1399.

MURPHY, T. 2015. Vaccines for Nontypeable Haemophilus influenzae: the Future Is Now. Clin Vaccine Immunol. , 22, 459-466.

MURPHY, T., BRAUER, A., SETHI, S., KILIAN, M., CAI, X. & LESSE, A. 2007. Haemophilus haemolyticus: a human respiratory tract commensal to be distinguished from Haemophilus influenzae. J Infect Dis. , 195, 81-89.

MURPHY, T. & PARAMESWARAN, G. 2009. Moraxella catarrhalis, a human respiratory tract pathogen. Clin Infect Dis., 49, 124-131.

MURPHY, T. F. 2005. Vaccine development for non-typeable Haemophilus influenzae and Moraxella catarrhalis: progress and challenges. Expert Rev. Vaccines, 4, 843-853.

MURRAY, J., MOORE, S., VAN DYKE, C., LAHR, B., DIERKHISING, R., ZINSMEISTER, A., MELTON, L., KRONING, C., EL-YOUSSEFF, M. & CZAJA, A. 2007. HLA DQ gene dosage and risk and severity of celiac disease. Clin Gastroenterol Hepatol., 5, 1406-1412.

NEUHAUSEN, S., FEOLO, M., CAMP, N., FARNHAM, J., BOOK, L. & ZONE, J. 2002. Genome-wide linkage analysis for celiac disease in North American families. Am J Med Genet., 111, 1-9.

NEWALL, A. T., CREIGHTON, P., PHILP, D. J., WOOD, J. G. & MACINTYRE, C. R. 2011. The potential cost-effectiveness of infant pneumococcal vaccines in Australia. Vaccine, 29, 8077-8085.

NGO, C., MASSA, H., THORNTON, R. & CRIPPS, A. 2016. Predominant Bacteria Detected from the Middle Ear Fluid of Children Experiencing Otitis Media: A Systematic Review. PloS One, 11, e0150949.

NGUYEN, K., HANSEN, E. & FARINHA, M. 1999. Construction of a genomic map of Moraxella (Branhamella) catarrhalis ATCC 25238 and physical mapping of virulence-associated genes. Can J Microbiol., 45, 299-303.

NISTICO, L., KREFT, R., GIESEKE, A., COTICCHIA, J., BURROWS, A., KHAMPANG, P., LIU, Y., KERSCHNER, J., POST, J., LONERGAN, S., SAMPATH, R., HU, F., EHRLICH, G., STOODLEY, P. & HALL-STOODLEY, L. 2011. Adenoid reservoir for pathogenic biofilm bacteria. J Clin Microbiol. , 49, 1411-1420.

257

NOKSO-KOIVISTO, J., CHONMAITREE, T., JENNINGS, K., MATALON, R., BLOCK, S. & PATEL, J. 2014. Polymorphisms of immunity genes and susceptibility to otitis media in children. PLoS One., 9, e93930.

NOKSO-KOIVISTO, J., HOVI, T. & PITKÄRANTA, A. 2006. Viral upper respiratory tract infections in young children with emphasis on acute otitis media. Int J Pediatr Otorhinolaryngol., 70, 1333-1342.

NOKSO-KOIVISTO, J., MAROM, T. & CHONMAITREE, T. 2015. Importance of viruses in acute otitis media. Curr Opin Pediatr., 27, 110-115.

NOKSO-KOIVISTO, J., RÄTY, R., BLOMQVIST, S., KLEEMOLA, M., SYRJÄNEN, R., PITKÄRANTA, A., KILPI, T. & HOVI, T. 2004. Presence of specific viruses in the middle ear fluids and respiratory secretions of young children with acute otitis media. J Med Virol. , 72, 241-248.

NORHAYATI, M., HO, J. & AZMAN, M. 2015. Influenza vaccine for preventing acute otitis media (middle ear infection). Cochrane Database Syst Rev., 3.

NUORTI, J. P. & WHITNEY, C. G. 2010. Prevention of Pneumococcal Disease Among Infants and Children — Use of 13-Valent Pneumococcal Conjugate Vaccine and 23-Valent Pneumococcal Polysaccharide Vaccine: Recommendations of the Advisory Committee on Immunization Practices (ACIP). Morbidity and Mortality Weekly Report (CDC), 59, 1-18.

O'BRIEN, M., PROSSER, L., PARADISE, J., RAY, G., KULLDORFF, M., KURS-LASKY, M., HINRICHSEN, V., MEHTA, J., COLBORN, D. & LIEU, T. 2009. New vaccines against otitis media: projected benefits and cost-effectiveness. Pediatrics., 123, 1452-1463.

O'CONNOR, T., PERRY, C. & LANNIGAN, F. 2009. Complications of otitis media in Indigenous and non-Indigenous children. Med J Aust., 191, S60-S64.

O'NEILL, P., ROBERTS, T. & BRADLEY STEVENSON, C. 2007. Otitis media in children (acute). BMJ Clin Evid., 2007.

OCHOA-GONDAR, O., FIGUEROLA-MASSANA, E., VILA-CORCOLES, A., AGUIRRE, C., DE DIEGO, C., SATUE, E., GOMEZ, F., RAGA, X. & EPIVAC STUDY GROUP 2015. Epidemiology of Streptococcus pneumoniae causing acute otitis media among children in Southern Catalonia throughout 2007-2013: Incidence, serotype distribution and vaccine's effectiveness. Int J Pediatr Otorhinolaryngol. , 79, 2104-2108.

OGRA, P. 2010. Ageing and its possible impact on mucosal immune responses. Ageing Res Rev. , 9, 101-106.

OPHIR, D., HAHN, T., SCHATTNER, A., WALLACH, D. & AVIEL, A. 1988. Tumor necrosis factor in middle ear effusions. Arch Otolaryngol Head Neck Surg., 114, 1256-1258.

OZAWA, D., YANO, H., ENDO, S., HIDAKA, H., KAKUTA, R., OKITSU, N., KANAMORI, H., OGAWA, M., ICHIMURA, S., SHIMOJIMA, M., INOMATA, S., KAKU, M. & KATORI, Y. 2015. Impact of the Seven-valent Pneumococcal Conjugate Vaccine on Acute Otitis Media in Japanese Children: Emergence of Serotype 15A Multidrug-resistant Streptococcus pneumoniae in Middle Ear Fluid Isolates. Pediatr Infect Dis J., 34, e217-e221.

OZGUR, S., BEYAZOVA, U., KEMALOGLU, Y., MARAL, I., SAHIN, F., CAMURDAN, A., KIZIL, Y., DINC, E. & TUZUN, H. 2006. Effectiveness of inactivated influenza vaccine for prevention of otitis media in children. Pediatr Infect Dis J., 25, 401-404.

PASSALI, D., PASSALI, G., LAURIELLO, M., ROMANO, A., BELLUSSI, L. & PASSALI, F. 2014. Nasal Allergy and Otitis Media: A real correlation? Sultan Qaboos Univ Med J. , 14, e59-e64.

PATEL, J., NAIR, S., GRADY, J., REVAI, K., VICTOR, S., BRASIER, A. & CHONMAITREE, T. 2009. Systemic cytokine response profiles associated with respiratory virus-induced acute otitis media. Pediatr Infect Dis J., 28, 407-411.

PATEL, J., NAIR, S., REVAI, K., GRADY, J., SAEED, K., MATALON, R., BLOCK, S. & CHONMAITREE, T. 2006. Association of proinflammatory cytokine gene polymorphisms with susceptibility to otitis media. Pediatrics., 118, 2273-2279.

258

PATEL, J. A., NGUYEN, D. T., REVAI, K. & CHONMAITREE, T. 2007. Role of respiratory syncytial virus in acute otitis media: implications for vaccine development. Vaccine, 25, 1683-1689.

PEREZ, A., PANG, B., KING, L., TAN, L., MURRAH, K., REIMCHE, J., WREN, J., RICHARDSON, S., GHANDI, U. & SWORDS, W. 2014. Residence of Streptococcus pneumoniae and Moraxella catarrhalis within polymicrobial biofilm promotes antibiotic resistance and bacterial persistence in vivo. Pathog Dis. , 70, 280-288.

PERKIN, R., SWIFT, J., NEWTON, D. & ANAS, N. 2007. Pediatric hospital medicine: textbook of inpatient management. 2nd ed.: Lippincott Williams & Wilkins.

PESTKA, S., KRAUSE, C., SARKAR, D., WALTER, M., SHI, Y. & FISHER, P. 2004. Interleukin-10 and related cytokines and receptors. Annu Rev Immunol., 22, 929-979.

PETTIGREW, M., GENT, J., PYLES, R., MILLER, A., NOKSO-KOIVISTO, J. & CHONMAITREE, T. 2011. Viral-bacterial interactions and risk of acute otitis media complicating upper respiratory tract infection. J Clin Microbiol., 49, 3750-3755.

PHILLIPS, B. & FLYNN, A. 2015. Novavax Initiates Global Pivotal Phase 3 Trial of the RSV F Vaccine to Protect Infants via Maternal Immunization [Online]. Novavax.

PICHICHERO, M., KAUR, R., CASEY, J., SABIROV, A., KHAN, M. & ALMUDEVAR, A. 2010. Antibody response to Haemophilus influenzae outer membrane protein D, P6, and OMP26 after nasopharyngeal colonization and acute otitis media in children. Vaccine., 28, 7184-7192.

PITKÄRANTA, A., HOVI, T. & KARMA, P. 1996. Interferon production by leukocytes in children with otitis media with effusion. Int J Pediatr Otorhinolaryngol., 34, 25-33.

PITKÄRANTA, A., JERO, J., ARRUDA, E., VIROLAINEN, A. & HAYDEN, F. 1998. Polymerase chain reaction-based detection of rhinovirus, respiratory syncytial virus, and coronavirus in otitis media with effusion. J Pediatr., 133, 390-394.

POST, J., AUL, J., WHITE, G., WADOWSKY, R., ZAVORAL, T., TABARI, R., KERBER, B., DOYLE, W. & EHRLICH, G. 1996. PCR-based detection of bacterial DNA after antimicrobial treatment is indicative of persistent, viable bacteria in the chinchilla model of otitis media. Am J Otolaryngol., 17, 106-111.

POST, J., HILLER, N., NISTICO, L., STOODLEY, P. & EHRLICH, G. 2007. The role of biofilms in otolaryngologic infections: update 2007. Curr Opin Otolaryngol Head Neck Surg., 15, 347-351.

POST, J. C., PRESTON, R. A., AUL, J. J., LARKINS-PETTIGREW, M., RYDQUIST-WHITE, J., ANDERSON, K. W. & AL, E. 1995. Molecular analysis of bacterial pathogens in otitis media with effusion. JAMA, 273, 1598-1604.

PRELLNER, K., KALM, O., HARSTEN, G., HELDRUP, J. & OXELIUS, V. 1989. Pneumococcal serum antibody concentrations during the first three years of life: a study of otitis-prone and non-otitis-prone children. Int J Pediatr Otorhinolaryngol., 17, 267-279.

PRICE, E., SAROVICH, D., NOSWORTHY, E., BEISSBARTH, J., MARSH, R., PICKERING, J., KIRKHAM, L., KEIL, A., CHANG, A. & SMITH-VAUGHAN, H. 2015. Haemophilus influenzae: using comparative genomics to accurately identify a highly recombinogenic human pathogen. BMC Genomics., 16.

PRYMULA, R., PEETERS, P., CHROBOK, V., KRIZ, P., NOVAKOVA, E., KALISKOVA, E., KOHL, I., LOMMEL, P., POOLMAN, J., PRIEELS, J. & SCHUERMAN, L. 2006. Pneumococcal capsular polysaccharides conjugated to protein D for prevention of acute otitis media caused by both Streptococcus pneumoniae and non-typable Haemophilus influenzae: a randomised double-blind efficacy study. Lancet, 367, 740-748.

PUMAROLA, F., MARÈS, J., LOSADA, I., MINGUELLA, I., MORAGA, F., TARRAGÓ, D. & ET AL 2013. Microbiology of bacteria causing recurrent acute otitis media (AOM) and AOM treatment failure in young children in Spain: shifting pathogens in the post-pneumococcal conjugate vaccination era. Int J Pediatr Otorhinolaryngol. , 77, 1231-1236.

259

RAWLINGS, B., HIGGINS, T. & HAN, J. 2013. Bacterial pathogens in the nasopharynx, nasal cavity, and osteomeatal complex during wellness and viral infection. Am J Rhinol Allergy. , 27, 39-42.

REN, D., ALMUDEVAR, A., MURPHY, T., LAFONTAINE, E., CAMPAGNARI, A., LUKE-MARSHALL, N., CASEY, J. & PICHICHERO, M. 2015. Serum antibody response to Moraxella catarrhalis proteins OMP CD, OppA, Msp22, Hag, and PilA2 after nasopharyngeal colonization and acute otitis media in children. Vaccine., 33, 5809-5814.

REN, D. & PICHICHERO, M. 2016. Vaccine targets against Moraxella catarrhalis. Expert Opin Ther Targets., 20, 19-33.

REVAI, K., PATEL, J., GRADY, J., NAIR, S., MATALON, R. & CHONMAITREE, T. 2009. Association between cytokine gene polymorphisms and risk for upper respiratory tract infection and acute otitis media. Clin Infect Dis., 49, 257-261.

RIHKANEN, H., CARPÉN, O., ROIVAINEN, M., VAHERI, A. & PITKÄRANTA, A. 2004. Rhinovirus in adenoid tissue. Int J Pediatr Otorhinolaryngol. , 68, 903-908.

ROCKETT, R., SLOOTS, T., BOWES, S., O'NEILL, N., YE, S., ROBSON, J., WHILEY, D., LAMBERT, S., WANG, D., NISSEN, M. & BIALASIEWICZ, S. 2013. Detection of novel polyomaviruses, TSPyV, HPyV6, HPyV7, HPyV9 and MWPyV in feces, urine, blood, respiratory swabs and cerebrospinal fluid. PLoS One., 8, e62764.

ROSENBLUT, A., SANTOLAYA, M. E., GONZALEZ, P., CORBALAN, V., AVENDANO, L. F., MARTINEZ, M. A. & HORMAZABAL, J. C. 2001. Bacterial and viral etiology of acute otitis media in Chilean children. The Pediatric Infectious Disease Journal, 20, 501-507.

ROSENFELD, R., CULPEPPER, L., DOYLE, K., GRUNDFAST, K., HOBERMAN, A., KENNA, M., LIEBERTHAL, A., MAHONEY, M., WAHL, R., WOODS, C. J., YAWN, B., AMERICAN ACADEMY OF PEDIATRICS SUBCOMMITTEE ON OTITIS MEDIA WITH EFFUSION, AMERICAN ACADEMY OF FAMILY PHYSICIANS & AMERICAN ACADEMY OF OTOLARYNGOLOGY--HEAD AND NECK SURGERY 2004. Clinical practice guideline: Otitis media with effusion. Otolaryngol Head Neck Surg., 130, S95-S118.

ROSENFELD, R. & KAY, D. 2003. Natural history of untreated otitis media. Laryngoscope., 113, 1645-1657.

ROSENFELD, R., SCHWARTZ, S., PYNNONEN, M., TUNKEL, D., HUSSEY, H., FICHERA, J., GRIMES, A., HACKELL, J., HARRISON, M., HASKELL, H., HAYNES, D., KIM, T., LAFRENIERE, D., LEBLANC, K., MACKEY, W., NETTERVILLE, J., PIPAN, M., RAOL, N. & SCHELLHASE, K. 2013. Clinical practice guideline: Tympanostomy tubes in children. Otolaryngol Head Neck Surg., 149, S1-S35.

ROSENFELD, R., SHIN, J., SCHWARTZ, S., COGGINS, R., GAGNON, L., HACKELL, J., HOELTING, D., HUNTER, L., KUMMER, A., PAYNE, S., POE, D., VELING, M., VILA, P., WALSH, S. & CORRIGAN, M. 2016. Clinical Practice Guideline: Otitis Media with Effusion (Update). Otolaryngol Head Neck Surg. , 154, S1-S41.

ROSENFELD, R., VERTREES, J., CARR, J., CIPOLLE, R., UDEN, D., GIEBINK, G. & CANAFAX, D. 1994. Clinical efficacy of antimicrobial drugs for acute otitis media: metaanalysis of 5400 children from thirty-three randomized trials. J Pediatr., 124, 355-367.

ROUT, M., MOHANTY, D., VIJAYLAXMI, Y., BOBBA, K. & METTA, C. 2013. Adenoid Hypertrophy in Adults: A case Series. Indian J Otolaryngol Head Neck Surg., 65, 269-274.

ROVERS, M., GLASZIOU, P., APPELMAN, C., BURKE, P., MCCORMICK, D., DAMOISEAUX, R., GABOURY, I., LITTLE, P. & HOES, A. 2006. Antibiotics for acute otitis media: a meta-analysis with individual patient data. Lancet., 368, 1429-1435.

ROVERS, M., SCHILDER, A., ZIELHUIS, G. & ROSENFELD, R. 2004. Otitis media. Lancet, 363, 465-473.

ROVERS, M. M. 2008. The burden of otitis media. Vaccine, 26, G2-G4. RUCKDESCHEL, E., BRAUER, A., JOHNSON, A. & MURPHY, T. 2009. Characterization of proteins

Msp22 and Msp75 as vaccine antigens of Moraxella catarrhalis. Vaccine., 27, 7065-7072.

260

RUOHOLA, A., MEURMAN, O., NIKKARI, S., SKOTTMAN, T., SALMI, A., WARIS, M., OSTERBACK, R., EEROLA, E., ALLANDER, T., NIESTERS, H., HEIKKINEN, T. & RUUSKANAN, O. 2006. Microbiology of acute otitis media in children with tympanostomy tubes: prevalences of bacteria and viruses. Clin Infect Dis. , 43, 1417-1422.

RUOHOLA, A., PETTIGREW, M., LINDHOLM, L., JALAVA, J., RÄISÄNEN, K., VAINIONPÄÄ, R., WARIS, M., TÄHTINEN, P., LAINE, M., LAHTI, E., RUUSKANEN, O. & HUOVINEN, P. 2013. Bacterial and viral interactions within the nasopharynx contribute to the risk of acute otitis media. J Infect., 66, 247-254.

RYANNA, K., STRATIGOU, V., SAFINIA, N. & HAWRYLOWICZ, C. 2009. Regulatory T cells in bronchial asthma. Allergy. , 64, 335-347.

RYE, M., WIERTSEMA, S., SCAMAN, E., OOMMEN, J., SUN, W., FRANCIS, R., ANG, W., PENNELL, C., BURGNER, D., RICHMOND, P., VIJAYASEKARAN, S., COATES, H., BROWN, S., BLACKWELL, J. & JAMIESON, S. 2011. FBXO11, a regulator of the TGFβ pathway, is associated with severe otitis media in Western Australian children. Genes Immun., 12, 352-359.

SAAFAN, M., IBRAHIM, W. & TOMOUM, M. 2013. Role of adenoid biofilm in chronic otitis media with effusion in children. Eur Arch Otorhinolaryngol., 270, 2417-2425.

SAKAGUCHI, S., WING, K., ONISHI, Y., PRIETO-MARTIN, P. & YAMAGUCHI, T. 2009. Regulatory T cells: how do they suppress immune responses? Int Immunol., 21, 1105-1111.

SANDERS, S., GLASZIOU, P., DEL MAR, C. & ROVERS, M. 2009. Antibiotics for acute otitis media in children (review). Cochrane database of systematic reviews, 2, 1-43.

SATO, M., LI, H., IKIZLER, M., WERKHAVEN, J., WILLIAMS, J., CHAPPELL, J., TANG, Y. & WRIGHT, P. 2009. Detection of viruses in human adenoid tissues by use of multiplex PCR. J Clin Microbiol. , 47, 771-773.

SATTUI, S., DE LA FLOR, C., SANCHEZ, C., LEWIS, D., LOPEZ, G., RIZO-PATRÓN, E., WHITE, A. J. & MONTES, M. 2012. Cryopreservation modulates the detection of regulatory T cell markers. Cytometry B Clin Cytom., 82, 54-58.

SAYLAM, G., TATAR, E., TATAR, I., OZDEK, A. & KORKMAZ, H. 2010. Association of adenoid surface biofilm formation and chronic otitis media with effusion. Arch Otolaryngol Head Neck Surg., 136, 550-555.

SCADDING, G. 2010. Non-surgical treatment of adenoidal hypertrophy: the role of treating IgE-mediated inflammation. Pediatr Allergy Immunol., 21, 1095-1106.

SCHARER, G., ZALDIVAR, F., GONZALEZ, G., VARGAS-SHIRAISHI, O., SINGH, J. & ARRIETA, A. 2003. Systemic inflammatory responses in children with acute otitis media due to Streptococcus pneumoniae and the impact of treatment with clarithromycin. Clin Diagn Lab Immunol. , 10, 721-724.

SEKIYAMA, K., OHORI, J., MATSUNE, S. & KURONO, Y. 2011. The role of vascular endothelial growth factor in pediatric otitis media with effusion. Auris Nasus Larynx., 38, 319-324.

SHARMA, S., CASEY, J. & PICHICHERO, M. 2012. Reduced serum IgG responses to pneumococcal antigens in otitis-prone children may be due to poor memory B-cell generation. J Infect Dis., 205, 1225-1229.

SHAW, C., OBERMYER, N., WETMORE, S., SPIROU, G. & FARR, R. 1995. Incidence of adenovirus and respiratory syncytial virus in chronic otitis media with effusion using the polymerase chain reaction. Otolaryngol Head Neck Surg., 113, 234-241.

SHINOGAMI, M. & ISHIBASHI, T. 2004. Presence of human herpesviruses in young children with acute otitis media. Int J Pediatr Otorhinolaryngol. , 68, 205-210.

SHIRAGAMI, M., MIZUKAMI, A., LEEUWENKAMP, O., MRKVAN, T., DELGLEIZE, E., KURONO, Y. & IWATA, S. 2015. Cost-Effectiveness Evaluation of the 10-Valent Pneumococcal Non-typeable Haemophilus influenzae Protein D Conjugate Vaccine and 13-Valent Pneumococcal Vaccine in Japanese Children. Infect Dis Ther., 4, 93-112.

261

SHORT, K., DIAVATOPOULOS, D., THORNTON, R., PEDERSEN, J., STRUGNELL, R., WISE, A., READING, P. & WIJBURG, O. 2011. Influenza virus induces bacterial and nonbacterial otitis media. J Infect Dis., 204, 1857-1865.

SILHAVY, T., KAHNE, D. & WALKER, S. 2010. The bacterial cell envelope. Cold Spring Harb Perspect Biol., 2, a000414.

SIMELL, B., KILPI, T. & KÄYHTY, H. 2002. Pneumococcal carriage and otitis media induce salivary antibodies to pneumococcal capsular polysaccharides in children. J Infect Dis., 186, 1106-1114.

SIMELL, B., MELIN, M., LAHDENKARI, M., BRILES, D., HOLLINGSHEAD, S., KILPI, T. & KAYHTY, H. 2007. Antibodies to pneumococcal surface protein A families 1 and 2 in serum and saliva of children and the risk of pneumococcal acute otitis media. J Infect Dis. , 196, 1528-1536.

SKOTNICKA, B. & HASSMANN, E. 2000. Cytokines in children with otitis media with effusion. Eur Arch Otorhinolaryngol., 257, 323-326.

SKOTNICKA, B. & HASSMANN, E. 2008. Proinflammatory and immunoregulatory cytokines in the middle ear effusions. Int J Pediatr Otorhinolaryngol., 72, 13-17.

SKOVSTED, I. 2015. Textbook in Diagnosis, Serotyping, Virulence Factors and Enzyme-linked Immunosorbent Assay (ELISA) for Measuring Pneumococcal Antibodies. In: ELVERDAL, P., KERRN, M., SØRENSEN, J., OTTE, S., FUURSTED, K., HARBOE, Z. & KONRADSEN, H. (eds.) Streptococcus pneumoniae. 2 ed. Denmark: Statens Serum Institut.

SLOYER, J. J., CATE, C., HOWIE, V., PLOUSSARD, J. & JOHNSTON, R. J. 1975. The immune response to acute otitis media in children. II. Serum and middle ear fluid antibody in otitis media due to Haemophilus influenza. J Infect Dis., 132, 685-688.

SLOYER, J. J., HOWIE, V., PLOUSSARD, J., SCHIFFMAN, G. & JOHNSTON, R. J. 1976. Immune response to acute otitis media: association between middle ear fluid antibody and the clearing of clinical infection. J Clin Microbiol. , 4.

SMIRNOVA, M., BIRCHALL, J. & PEARSON, J. 2005. Evidence of T-helper cell 2 cytokine regulation of chronic otitis media with effusion. Acta Otolaryngol., 125, 1043-1050.

SMIRNOVA, M., KISELEV, S., GNUCHEV, N., BIRCHALL, J. & PEARSON, J. 2002. Role of the pro-inflammatory cytokines tumor necrosis factor-alpha, interleukin-1 beta, interleukin-6 and interleukin-8 in the pathogenesis of the otitis media with effusion. Eur Cytokine Netw., 13, 161-172.

SMITH-VAUGHAN, H., BYUN, R., HALPIN, S., NADKARNI, M., JACQUES, N., HUNTER, N., MORRIS, P. & LEACH, A. 2008. Interventions for prevention of otitis media may be most effective if implemented in the first weeks of life. Int J Pediatr Otorhinolaryngol. , 72, 57-61.

SMITH-VAUGHAN, H., BYUN, R., NADKARNI, M., JACQUES, N., HUNTER, N., HALPIN, S., MORRIS, P. & LEACH, A. 2006. Measuring nasal bacterial load and its association with otitis media. BMC Ear Nose Throat Disord., 6.

SMITH-VAUGHAN, H. C., CHANG, A. B., SAROVICH, D. S., MARSH, R. L., GRIMWOOD, K., LEACH, A. J., MORRIS, P. S. & PRICE, E. P. 2014. Absence of an important vaccine and diagnostic target in carriage- and disease-related nontypeable Haemophilus influenzae. Clin Vaccine Immunol., 21, 250-252.

SOCIETY FOR MIDDLE EAR DISEASE 2012. Viewed 15 March 2012 from. https://societyformiddleeardisease.org/faq/.

SØRENSEN, C. & NIELSEN, L. 1988. Nasopharyngeal secretory immunoglobulins in children with recurrent acute otitis media and secretory otitis media. APMIS., 96, 199-205.

SPEEL, E. 1999. Robert Feulgen Prize Lecture 1999. Detection and amplification systems for sensitive, multiple-target DNA and RNA in situ hybridization: looking inside cells with a spectrum of colors. Histochem Cell Biol., 112, 89-113.

SPIJKERMAN, J., PREVAES, S., VAN GILS, E., VEENHOVEN, R., BRUIN, J., BOGAERT, D., WIJMENGA-MONSUUR, A., VAN DEN DOBBELSTEEN, G. & SANDERS, E. 2012. Long-term effects of

https://societyformiddleeardisease.org/faq/

262

pneumococcal conjugate vaccine on nasopharyngeal carriage of S. pneumoniae, S. aureus, H. influenzae and M. catarrhalis. PLoS One., 7, e39730.

STOL, K., DIAVATOPOULOS, D., GRAAMANS, K., ENGEL, J., MELCHERS, W., SAVELKOUL, H., HAYS, J., WARRIS, A. & HERMANS, P. 2012. Inflammation in the middle ear of children with recurrent or chronic otitis media is associated with bacterial load. Pediatr Infect Dis J., 31, 1128-1134.

STOL, K., VERHAEGH, S. J., GRAAMANS, K., ENGEL, J. A., STURM, P. D., MELCHERS, W. J. & AT, E. 2013. Microbial profiling does not differentiate between childhood recurrent acute otitis media and chronic otitis media with effusion. Int J Pediatr Otorhinolaryngol. , 77, 488-493.

STOODLEY, P. & DIRCKX, P. 2003. Viewed 22 March 2016 from. http://www.biofilm.montana.edu/resources/images/multicellularextracellular/biofilm-formation-3-steps.html.

STORGAARD, M., LARSEN, K., BLEGVAD, S., NØDGAARD, H., OVESEN, T., ANDERSEN, P. & OBEL, N. 1997. Interleukin-8 and chemotactic activity of middle ear effusions. J Infect Dis. , 175, 474-477.

SUN, W., JACOBY, P., RILEY, T., BOWMAN, J., LEACH, A., COATES, H., WEEKS, S., CRIPPS, A., LEHMANN, D. & KALGOORLIE OTITIS MEDIA RESEARCH PROJECT TEAM 2012. Association between early bacterial carriage and otitis media in Aboriginal and non-Aboriginal children in a semi-arid area of Western Australia: a cohort study. BMC Infect Dis., 12.

SURENDRAN, N., NICOLOSI, T., KAUR, R. & PICHICHERO, M. 2016. Peripheral blood antigen presenting cell responses in otitis-prone and non-otitis-prone infants. Innate Immun., 22, 63-71.

SUZUKI, K. & BAKALETZ, L. 1994. Synergistic effect of adenovirus type 1 and nontypeable Haemophilus influenzae in a chinchilla model of experimental otitis media. Infect Immun., 62, 1710-1718.

SUZUKI, K., SETOYAMA, Y., YOSHIMOTO, K., TSUZAKA, K., ABE, T. & TAKEUCHI, T. 2011. Decreased mRNA expression of two FOXP3 isoforms in peripheral blood mononuclear cells from patients with rheumatoid arthritis and systemic lupus erythematosus. Int J Immunopathol Pharmacol., 24, 7-14.

SUZUKI, M., WATANABE, T. & MOGI, G. 1999. Clinical, bacteriological, and histological study of adenoids in children. Am J Otolaryngol. , 20, 85-90.

SWARTS, J., ALPER, C., LUNTZ, M., BLUESTONE, C., DOYLE, W., GHADIALI, S., POE, D., TAKAHASHI, H. & TIDEHOLM, B. 2013. Panel 2: Eustachian tube, middle ear, and mastoid--anatomy, physiology, pathophysiology, and pathogenesis. Otolaryngol Head Neck Surg. , 148, E26-36.

SZALMÁS, A., PAPP, Z., CSOMOR, P., KÓNYA, J., SZIKLAI, I., SZEKANECZ, Z. & KAROSI, T. 2013. Microbiological profile of adenoid hypertrophy correlates to clinical diagnosis in children. Biomed Res Int., 2013.

TÄHTINEN, P., LAINE, M., HUOVINEN, P., JALAVA, J., RUUSKANEN, O. & RUOHOLA, A. 2011. A placebo-controlled trial of antimicrobial treatment for acute otitis media. N Engl J Med. , 364, 116-126.

TAKEUCHI, K., MAESAKO, K., YUTA, A. & SAKAKURA, Y. 1994. Interleukin-8 gene expression in middle ear effusions. Ann Otol Rhinol Laryngol. , 103, 404-407.

TAMIR, S., ROTH, Y., DALAL, I., GOLDFARB, A., GROTTO, I. & MAROM, T. 2015. Changing trends of acute otitis media bacteriology in central Israel in the pneumococcal conjugate vaccines era. Pediatr Infect Dis J., 34, 195-199.

TAN, T. 2012. Pediatric invasive pneumococcal disease in the United States in the era of pneumococcal conjugate vaccines. Clin Microbiol Rev., 25, 409-419.

http://www.biofilm.montana.edu/resources/images/multicellularextracellular/biofilm-formation-3-steps.html

http://www.biofilm.montana.edu/resources/images/multicellularextracellular/biofilm-formation-3-steps.html

263

TAYLOR, P., FAETH, I., MARKS, M., DEL MAR, C., SKULL, S., PEZZULLO, M., HAVYATT, S. & COATES, H. 2009. Cost of treating otitis media in Australia. Expert Rev Pharmacoecon Outcomes Res., 9, 133-141.

THORNTON, R. B., RIGBY, P. J., WIERTSEMA, S. P., FILION, P., LANGLANDS, J., COATES, H. L., VIJAYASEKARAN, S., KEIL, A. D. & RICHMOND, P. C. 2011. Multi-species bacterial biofilm and intracellular infection in otitis media. BMC Pediatr., 11, 1-10.

THORNTON, R. B., WIERTSEMA, S. P., KIRKHAM, L. S., RIGBY, P. J., VIJAYASEKARAN, S., COATES, H. L. & RICHMOND, P. C. 2013. Neutrophil extracellular traps and bacterial biofilms in middle ear effusion of children with recurrent acute otitis media--a potential treatment target. PLoS One, 8, e53837.

TOMONAGA, K., KURONO, Y., CHAEN, T. & MOGI, G. 1989. Adenoids and otitis media with effusion: nasopharyngeal flora. Am J Otolaryngol. , 10, 204-207.

TOZER, S., LAMBERT, S., WHILEY, D., BIALASIEWICZ, S., LYON, M., NISSEN, M. & SLOOTS, T. 2009. Detection of human bocavirus in respiratory, fecal, and blood samples by real-time PCR. J Med Virol., 81, 488-493.

TREGNAGHI, M., SÁEZ-LLORENS, X., LÓPEZ, P., ABATE, H., SMITH, E., PÓSLEMAN, A., CALVO, A., WONG, D., CORTES-BARBOSA, C., CEBALLOS, A., TREGNAGHI, M., SIERRA, A., RODRIGUEZ, M., TROITIÑO, M., CARABAJAL, C., FALASCHI, A., LEANDRO, A., CASTREJÓN, M., LEPETIC, A., LOMMEL, P., HAUSDORFF, W., BORYS, D., RUIZ GUIÑAZÚ, J., ORTEGA-BARRÍA, E., YARZÁBAL, J., SCHUERMAN, L. & COMPAS GROUP 2014. Efficacy of pneumococcal nontypable Haemophilus influenzae protein D conjugate vaccine (PHiD-CV) in young Latin American children: A double-blind randomized controlled trial. PLoS Med., 11, e1001657.

TYO, K. R., ROSEN, M. M., ZENG, W., YAP, M., PWEE, K. H., ANG, L. W. & SHEPARD, D. S. 2011. Cost-effectiveness of conjugate pneumococcal vaccination in Singapore: comparing estimates for 7-valent, 10-valent, and 13-valent vaccines. Vaccine, 29, 6686-6694.

VAN DE VOOREN, K., DURANTI, S., CURTO, A. & GARATTINI, L. 2014. Cost effectiveness of the new pneumococcal vaccines: a systematic review of European studies. Pharmacoeconomics., 32, 29-45.

VAN DEN AARDWEG, M., SCHILDER, A., HERKERT, E., BOONACKER, C. & ROVERS, M. 2010. Adenoidectomy for otitis media in children. Cochrane Database Syst Rev., 1.

VAN ELDEN, L., VAN LOON, A., VAN ALPHEN, F., HENDRIKSEN, K., HOEPELMAN, A., VAN KRAAIJ, M., OOSTERHEERT, J., SCHIPPER, P., SCHUURMAN, R. & NIJHUIS, M. 2004. Frequent detection of human coronaviruses in clinical specimens from patients with respiratory tract infection by use of a novel real-time reverse-transcriptase polymerase chain reaction. J Infect Dis., 189, 652-657.

VAN KEMPEN, M., VERMEIREN, J., VANEECHOUTTE, M., CLAEYS, G., VEENHOVEN, R., RIJKERS, G., SANDERS, E. & DHOOGE, I. 2006. Pneumococcal conjugate vaccination in children with recurrent acute otitis media: a therapeutic alternative? Int J Pediatr Otorhinolaryngol. , 70, 275-285.

VAN ZON, A., VAN DER HEIJDEN, G., VAN DONGEN, T., BURTON, M. & SCHILDER, A. 2012. Antibiotics for otitis media with effusion in children. Cochrane Database Syst Rev. , 9.

VAREILLE, M., KIENINGER, E., EDWARDS, M. & REGAMEY, N. 2011. The airway epithelium: soldier in the fight against respiratory viruses. Clin Microbiol Rev. , 24, 210-229.

VEENHOVEN, R., RIJKERS, G., SCHILDER, A., ADELMEIJER, J., UITERWAAL, C., KUIS, W. & SANDERS, E. 2004. Immunoglobulins in otitis-prone children. Pediatr Res., 55, 159-162.

VERHAEGH, S., STOL, K., DE VOGEL, C., RIESBECK, K., LAFONTAINE, E., MURPHY, T., VAN BELKUM, A., HERMANS, P. & HAYS, J. 2012. Comparative analysis of the humoral immune response to Moraxella catarrhalis and Streptococcus pneumoniae surface antigens in children suffering from recurrent acute otitis media and chronic otitis media with effusion. Clin Vaccine Immunol. , 19, 914-918.

264

VESIKARI, T., FORSTEN, A., SEPPÄ, I., KAIJALAINEN, T., PUUMALAINEN, T., SOININEN, A., TRASKINE, M., LOMMEL, P., SCHOONBROODT, S., HEZAREH, M., MOREIRA, M., BORYS, D. & SCHUERMAN, L. 2016. Effectiveness of the 10-Valent Pneumococcal Nontypeable Haemophilus influenzae Protein D-Conjugated Vaccine (PHiD-CV) Against Carriage and Acute Otitis Media-A Double-Blind Randomized Clinical Trial in Finland. J Pediatric Infect Dis Soc.

VIROLAINEN, A., JERO, J., KÄYHTY, H., KARMA, P., LEINONEN, M. & ESKOLA, J. 1995. Antibodies to pneumolysin and pneumococcal capsular polysaccharides in middle ear fluid of children with acute otitis media. Acta Otolaryngol., 115, 796-803.

WANG, Y., KELLY, C., KARTTUNEN, J., WHITTALL, T., LEHNER, P., DUNCAN, L., MACARY, P., YOUNSON, J., SINGH, M., OEHLMANN, W., CHENG, G., BERGMEIER, L. & LEHNER, T. 2001. CD40 is a cellular receptor mediating mycobacterial heat shock protein 70 stimulation of CC-chemokines. Immunity. , 15, 971-983.

WATSON, K., CARVILLE, K., BOWMAN, J., JACOBY, P., RILEY, T., LEACH, A., LEHMANN, D. & TEAM., K. O. M. R. P. 2006. Upper respiratory tract bacterial carriage in Aboriginal and non-Aboriginal children in a semi-arid area of Western Australia. Pediatr Infect Dis J. , 25, 782-790.

WEIL-OLIVIER, C., VAN DER LINDEN, M., DE SCHUTTER, I., DAGAN, R. & MANTOVANI, L. 2012. Prevention of pneumococcal diseases in the post-seven valent vaccine era: a European perspective. BMC Infect Dis, 12, 1-12.

WEISS, S., FALKENHORST, G., VAN DER LINDEN, M., IMOHL, M. & VON KRIES, R. 2015. Impact of 10- and 13-valent pneumococcal conjugate vaccines on incidence of invasive pneumococcal disease in children aged under 16 years in Germany, 2009 to 2012. Euro Surveill., 20, 21057.

WESTERBERG, B., KOZAK, F., THOMAS, E., BLONDEL-HILL, E., BRUNSTEIN, J. & PATRICK, D. 2009. Is the healthy middle ear a normally sterile site? Otol Neurotol., 30, 174-177.

WHILEY, D. & SLOOTS, T. 2005. A 5'-nuclease real-time reverse transcriptase-polymerase chain reaction assay for the detection of a broad range of influenza A subtypes, including H5N1. Diagn Microbiol Infect Dis., 53, 335-337.

WIERTSEMA, S., CHIDLOW, G., KIRKHAM, L., CORSCADDEN, K., MOWE, E., VIJAYASEKARAN, S., COATES, H., HARNETT, G. & RICHMOND, P. 2011a. High detection rates of nucleic acids of a wide range of respiratory viruses in the nasopharynx and the middle ear of children with a history of recurrent acute otitis media. J Med Virol., 83, 2008-2017.

WIERTSEMA, S., CORSCADDEN, K., MOWE, E., ZHANG, G., VIJAYASEKARAN, S., COATES, H., MITCHELL, T., THOMAS, W., RICHMOND, P. & KIRKHAM, L. 2012. IgG responses to Pneumococcal and Haemophilus influenzae protein antigens are not impaired in children with a history of recurrent acute otitis media. PLoS One. , 7, e49061.

WIERTSEMA, S. & LEACH, A. 2009. Theories of otitis media pathogenesis, with a focus on Indigenous children. Med J Aust. , 191, S50-S54.

WIERTSEMA, S. P., KIRKHAM, L. S., CORSCADDEN, K. J., MOWE, E. N., BOWMAN, J. M., JACOBY, P., FRANCIS, R., VIJAYASEKARAN, S., COATES, H., RILEY, T. V. & RICHMOND, P. 2011b. Predominance of nontypeable Haemophilus influenzae in children with otitis media following introduction of a 3 + 0 pneumococcal conjugate vaccine schedule. Vaccine, 29, 5163-5170.

WINTHER, B., ALPER, C., MANDEL, E., DOYLE, W. & HENDLEY, J. 2007. Temporal relationships between colds, upper respiratory viruses detected by polymerase chain reaction, and otitis media in young children followed through a typical cold season. Pediatrics. , 119, 1069-1075.

WINTHER, B., HAYDEN, F. & HENDLEY, J. 2006. Picornavirus infections in children diagnosed by RT-PCR during longitudinal surveillance with weekly sampling: Association with symptomatic illness and effect of season. J Med Virol. , 78, 644-650.

265

WISLSON, M. 2005. Microbial inhabitants of human: their ecology and role in health and disease, Cambridge University Press, Cambridge, United Kingdom.

WRIGHT, E., HURST, D., MIOTTO, D., GIGUERE, C. & HAMID, Q. 2000. Increased expression of major basic protein (MBP) and interleukin-5(IL-5) in middle ear biopsy specimens from atopic patients with persistent otitis media with effusion. Otolaryngol Head Neck Surg. , 123, 533-538.

WROBLEWSKI, D., HALSE, T., HAYES, J., KOHLERSCHMIDT, D. & MUSSER, K. 2013. Utilization of a real-time PCR approach for Haemophilus influenzae serotype determination as an alternative to the slide agglutination test. Mol Cell Probes., 27, 86-89.

WU, S., LI, X., GUNAWARDANA, M., MAGUIRE, K., GUERRERO-GIVEN, D., SCHAUDINN, C., WANG, C., BAUM, M. & WEBSTER, P. 2014. Beta- lactam antibiotics stimulate biofilm formation in non-typeable haemophilus influenzae by up-regulating carbohydrate metabolism. PLoS One, 9, e99204.

XU, Q., CASEY, J. & PICHICHERO, M. 2015. Higher levels of mucosal antibody to pneumococcal vaccine candidate proteins are associated with reduced acute otitis media caused by Streptococcus pneumoniae in young children. Mucosal Immunol., 8, 1110-1117.

YANO, H., OKITSU, N., HORI, T., WATANABE, O., KISU, T., HATAGISHI, E., SUZUKI, A., OKAMOTO, M., OHMI, A., SUETAKE, M., SAGAI, S., KOBAYASHI, T. & NISHIMURA, H. 2009a. Detection of respiratory viruses in nasopharyngeal secretions and middle ear fluid from children with acute otitis media. Acta Oto-Laryngologica, 129, 19-24.

YANO, H., OKITSU, N., HORI, T., WATANABE, O., KISU, T., HATAGISHI, E., SUZUKI, A., OKAMOTO, M., OHMI, A., SUETAKE, M., SAGAI, S., KOBAYASHI, T. & NISHIMURA, H. 2009b. Detection of respiratory viruses in nasopharyngeal secretions and middle ear fluid from children with acute otitis media. Acta Otolaryngol., 129, 19-24.

YELLON, R., DOYLE, W., WHITESIDE, T., DIVEN, W., MARCH, A. & FIREMAN, P. 1995. Cytokines, immunoglobulins, and bacterial pathogens in middle ear effusions. Arch Otolaryngol Head Neck Surg. , 121, 865-869.

YELLON, R., LEONARD, G., MARUCHA, P., CRAVEN, R., CARPENTER, R., LEHMANN, W., BURLESON, J. & KREUTZER, D. 1991. Characterization of cytokines present in middle ear effusions. Laryngoscope., 101, 165-169.

YELLON, R., LEONARD, G., MARUCHA, P., SIDMAN, J., CARPENTER, R., BURLESON, J., CARLSON, J. & KREUTZER, D. 1992. Demonstration of interleukin 6 in middle ear effusions. Arch Otolaryngol Head Neck Surg., 118, 745-748.

YUAN, C., MILEY, W. & WATERS, D. 2001. A quantification of human cells using an ERV-3 real time PCR assay. J Virol Methods., 91, 109-117.

YUAN, S., COHEN, D., RAVEL, J., ABDO, Z. & FORNEY, L. 2012. Evaluation of methods for the extraction and purification of DNA from the human microbiome. Plos One, 7, e33865.

ZHANG, Q., LEONG, S., MCNAMARA, P., MUBARAK, A., MALLEY, R. & FINN, A. 2011. Characterisation of regulatory T cells in nasal associated lymphoid tissue in children: relationships with pneumococcal colonization. PLoS Pathog., 7, e1002175.

ZHAO, S., LI, J., LIU, H., ZHANG, Q., WANG, Y. & HAN, D. 2009. Role of interleukin-10 and transforming growth factor beta 1 in otitis media with effusion. Chin Med J (Engl). 122, 2149-2154.

ZIELNIK-JURKIEWICZ, B. & STANKIEWICZ-SZYMCZAK, W. 2015. Pro-inflammatory interleukins in middle ear effusions from atopic and non-atopic children with chronic otitis media with effusion. Eur Arch Otorhinolaryngol. , [Epub ahead of print].

266

Appendix A:

Bacterial otopathogens: systemic review

RESEARCH ARTICLE

Predominant Bacteria Detected from theMiddle Ear Fluid of Children ExperiencingOtitis Media: A Systematic ReviewChinh C. Ngo1,2, Helen M. Massa1,2*, Ruth B. Thornton3,4, Allan W. Cripps1,2

1 School of Medical Science, Griffith University, Gold Coast, Queensland, Australia, 2 Molecular Basis ofDisease, Menzies Health Institute Queensland, Griffith University, Gold Coast, Queensland, Australia,3 School of Paediatrics and Child Health, University of Western Australia, Perth, Western Australia,Australia, 4 Telethon Kids Institute, University of Western Australia, Perth, Western Australia, Australia

* [email protected]

Abstract

Background

Otitis media (OM) is amongst the most common childhood diseases and is associated with

multiple microbial pathogens within the middle ear. Global and temporal monitoring of pre-

dominant bacterial pathogens is important to inform new treatment strategies, vaccine

development and to monitor the impact of vaccine implementation to improve progress

toward global OM prevention.

Methods

A systematic review of published reports of microbiology of acute otitis media (AOM) and

otitis media with effusion (OME) from January, 1970 to August 2014, was performed using

PubMed databases.

Results

This review confirmed that Streptococcus pneumoniae and Haemophilus influenzae,remain the predominant bacterial pathogens, with S. pneumoniae the predominant bacte-

rium in the majority reports from AOM patients. In contrast, H. influenzae was the predomi-

nant bacterium for patients experiencing chronic OME, recurrent AOM and AOM with

treatment failure. This result was consistent, even where improved detection sensitivity

from the use of polymerase chain reaction (PCR) rather than bacterial culture was con-

ducted. On average, PCR analyses increased the frequency of detection of S. pneumoniaeand H. influenzae 3.2 fold compared to culture, whilstMoraxella catarrhalis was 4.5 times

more frequently identified by PCR. Molecular methods can also improve monitoring of

regional changes in the serotypes and identification frequency of S. pneumoniae and H.influenzae over time or after vaccine implementation, such as after introduction of the 7-

valent pneumococcal conjugate vaccine.

PLOS ONE | DOI:10.1371/journal.pone.0150949 March 8, 2016 1 / 26

OPEN ACCESS

Citation: Ngo CC, Massa HM, Thornton RB, CrippsAW (2016) Predominant Bacteria Detected from theMiddle Ear Fluid of Children Experiencing OtitisMedia: A Systematic Review. PLoS ONE 11(3):e0150949. doi:10.1371/journal.pone.0150949

Editor: Sean Reid, Ross University School ofMedicine, DOMINICA

Received: August 9, 2015

Accepted: February 22, 2016

Published: March 8, 2016

Copyright: © 2016 Ngo et al. This is an open accessarticle distributed under the terms of the CreativeCommons Attribution License, which permitsunrestricted use, distribution, and reproduction in anymedium, provided the original author and source arecredited.

Data Availability Statement: All relevant data arewithin the paper, its references and its SupportingInformation files.

Funding: The authors received no specific fundingfor this work. CNN acknowledges the VietnamInternational Education Development, The Ministry ofEducation and Training (VIED-MOET) for a doctoralscholarship. RBTacknowledges the BrightSparkFoundation, Perth, Western Australia for a post-doctoral fellowship. The funders had no role in thestudy design, data collection and analysis, decision topublish, or preparation of the manuscript.

http://crossmark.crossref.org/dialog/?doi=10.1371/journal.pone.0150949&domain=pdf

http://creativecommons.org/licenses/by/4.0/

http://creativecommons.org/licenses/by/4.0/

Conclusions

Globally, S. pneumoniae and H. influenzae remain the predominant otopathogens associ-

ated with OM as identified through bacterial culture; however, molecular methods continue

to improve the frequency and accuracy of detection of individual serotypes. Ongoing moni-

toring with appropriate detection methods for OM pathogens can support development of

improved vaccines to provide protection from the complex combination of otopathogens

within the middle ear, ultimately aiming to reduce the risk of chronic and recurrent OM in vul-

nerable populations.

IntroductionOtitis media, may be simply defined as inflammation of the middle ear, and is the most com-mon reason a child under the age of 5 will visit their general practitioner and be prescribedantibiotics in socioeconomically developed countries [1]. OM has a range of clinical presenta-tions including AOM, which is characterised by the rapid onset of local and systemic symp-toms, including otalgia, fever, vomiting and accumulation of fluid in the middle ear cavity andOME, where the child experiences MEF accumulation without the systemic symptoms. Both ofthese presentations may occur recurrently or chronically [2]. Globally, more than 700 millioncases of AOM are diagnosed each year, with 50% of affected children being under five years ofage [3]. Recurrent AOM (RAOM) occurs where a patient has 3 diagnosed AOM episodeswithin six months or more than 4 episodes in 12 months [4] and is commonly observed in upto 65% of children by 5 years of age[4] OME typically resolves spontaneously within 3 months[5] however, 30–40% of children experience persistent or chronic fluid in the middle ear formore than 3 months (chronic OME, COME) and may require surgical intervention to aid reso-lution [5].

Irrespective of clinical presentation, OM is a multi-factorial disease, with many associatedrisk factors, including environmental, immunological deficiency, gender, age and microbialexposure [4, 6]. Despite this complexity, bacterial and viral pathogens, individually andtogether, are strongly associated with OM development, for example, only 4% children diag-nosed with AOM had no bacterial or viral pathogen detected using culture and PCR [7].

The three bacteria most frequently identified in association with OM are: S. pneumoniae,H.influenzae andM. catarrhalis [6], whilst the viruses most commonly associated with OM arerespiratory syncytial virus, adenovirus, rhinovirus and coronavirus [6].

Monitoring both the identification and frequency of detection of common otopathogens inOM is central to evaluation of the effects of treatment and impact of vaccination programs. Forexample, pneumococcal carriage was reduced in association with the introduction of the hepta-valent pneumococcal conjugate vaccine (PCV7) in the US and has also increased prevention ofearly AOM infections [8].

Implementation of PCV7 into national immunisation programs (NIPs) has also resulted inshifts to non-vaccine pneumococcal serotypes isolated from invasive pneumococcal diseaseand OM [9–12]. Importantly, PCV7 implementation has resulted in non-typeable (non-encap-sulated) H. influenzae (NTHi) detection frequency surpassing S. pneumoniae detection inAOM patients within a number of countries including the US [12], Spain [13] and France [14].The emergence of non-vaccine serotypes and their potential role as pathogens in OM is an areaof ongoing research interest.

Bacteria in the Middle Ear of Children with Otitis Media


Competing Interests: AWC received a researchgrant from Danisco A/S on the use of probiotics forrespiratory illness and a research grant andconsultancy from Probiotec Pty Ltd on the use ifnutritional supplements for immune health. Theothers authors declare that they have no competinginterests. This does not alter the authors' adherenceto PLOS ONE policies on sharing data and materials.

This systematic review examines the frequency of detection of bacteria within MEF fromchildren experiencing OM, globally. Ongoing surveillance of bacteria predominant in OMpatients, improved detection methods such as PCR and monitoring the impact of immunisa-tions will continue to inform treatment decisions and ultimately vaccine development aimed toprevent OM development in young children.

Methods

Data sources and searchesThe design and construction of this systematic review was informed and in compliance withPRISMA (2009) [15] (see S1 PRISMA Checklist). Published literature, including conferenceabstracts, was searched via PubMed database and only original articles with an English abstractwere retrieved and reviewed. Publication dates were restricted from January 1st, 1970 to August30th, 2014, with five different search strategies used, as detailed below.

PubMed search strategyThe following terms were used in five literature searches:

• Search 1—Asian region: (otitis media (Asia (aetiology, otopathogens, pathogens, microbiol-ogy, bacteriology, bacteria))) or (otitis media (Bangladesh, Brunei, Cambodia, China, India,Indonesia, Iran, Israel, Japan, Korea, Laos, Lebanon, Malaysia, Pakistan, Philippines, SaudiArabia, Singapore, Taiwan, Thailand, Turkey, Vietnam)) (S1 Fig)

• Search 2—American region: (otitis media (America (aetiology, otopathogens, pathogens,microbiology, bacteriology, bacteria))) or (otitis media (Argentina, Brazil, Chile, Colombia,Costa Rica, Cuba, Dominica Republic, Ecuador, Honduras, Mexico, Panama, Paraguay, Uru-guay, Venezuela, Canada, The US)) (S2 Fig)

• Search 3—African region: (otitis media (Africa (aetiology, otopathogens, pathogens, microbi-ology, bacteriology, bacteria))) or (otitis media (Algeria, Cameroon, Cote D’lvoire, Egypt,Mozambique, Namibia, Nigeria, South Africa, Zambia, Zimbabwe)) (S3 Fig)

• Search 4—European region: (otitis media (Europe (aetiology, otopathogens, pathogens,microbiology, bacteriology, bacteria))) or (otitis media (Britain, Denmark, England, France,Finland, Germany, Greece, Ireland, Italy, Netherlands, Norway, Poland, Portugal, Romania,Russia, Scotland, Spain, Sweden, Ukraine, Wales, The UK)) (S4 Fig)

• Search 5—Oceanian region: (otitis media (Oceania (aetiology, otopathogens, pathogens,microbiology, bacteriology, bacteria))) or (otitis media (Australia, Cook Islands, New Zea-land, Papua New Guinea, Solomon Islands)) (S5 Fig)

Selection criteriaAll articles in English were assessed for relevance by review of abstracts when available, prior tofull review of each article. Pre-determined inclusion selection criteria were used: (i) onlyhuman studies (ii) age range one month– 18 years (majority of reported study populationsbelow 8 years old) (iii) otopathogenic bacteria were identified fromMEF samples of childrenexperiencing a range of presentations of OM, including AOM/RAOM/AOMTF or OME/COME but not chronic suppurative OM (CSOM) (iv) MEF samples were collected throughclinical perforation of the tympanic membrane (myringotomy or tympanocentesis) only,rather than spontaneous perforation.



Study selection and data extractionAll articles meeting the keyword criteria, identified in the primary database search, werescreened independently for relevancy by two of the authors (C.C.N. and H.M.M) who read allarticle titles and abstracts in a standardised manner. Abstracts and articles were reviewed andcompared to the inclusion criteria above and key data evaluating the reported details of thestudy population (age), types of OM, sample size, sample collection methods, microbiologicalmethod and years of study were recorded for included studies. Furthermore, where the detailof the sample collection method or clinical criteria for diagnosis of type of OM was inconclu-sive, the full paper was retrieved and reviewed to determine inclusion. Discussion betweenreviewers resolved any disagreement regarding inclusion. Additional information was soughtfrom the authors of included reviews, via emailed request seeking additional unpublishedinformation regarding vaccination status and/or serotyping data, if available for the study pop-ulations, or to clarify information, such as sampling methods used in their published study. Allbibliographies of the included studies were manually searched for additional relevantreferences.

Data items, synthesis and analysisData illustrating the frequency of detection of predominant bacteria from each report were col-lated and recorded. The antibiotic sensitivity of these bacteria and the vaccination status of theparticipants were recorded, where available. Data tables within this narrative report were com-piled from the primary results from each included report. Meta-analysis was not performeddue to variation of the designs of the included studies and study population characteristicswithin the published reports.

Evaluation of study qualityIn this study, included studies were previously published in peer reviewed journals. The varietyof individual study designs prevented use of a validated instrument to assess the quality ofincluded studies. The assessment of risk of bias of individual studies was performed at thestudy level by the use of inclusion criteria. For example, clinical criteria describing the variedpresentations of OMmay have varied over the thirty years of published literature reviewed.Reviewing authors ensured that all published studies clearly differentiated acute AOM by thepresence of otalgia and inflamed and/or bulging tympanic membrane which remain consistentcharacteristics within current definitions. Where vaccination status of the study population isunknown, comparison of the study dates versus the implementation of the NIPs was used toinform potential vaccination status. Only studies that sampled the MEF after clinical perfora-tion of the tympanic membrane were included in this report. This criterion is likely to havegenerated selection bias on the basis of MEF collection being performed primarily for researchpurposes, rather than clinical or treatment reasons. The likely outcome would result in skewingrepresentation to include more severe cases undergoing surgical treatment. In addition, there isa relative paucity of reports from some world regions.

ResultsThe search “otitis media and children” identified 9,617 articles via PubMed database searchover the period January 1st, 1970 to August 30th, 2014. After performing the key word screeningfor each of the 5 regions, as illustrated in Fig 1, 10,483 articles were identified, screened for eli-gibility and 857 abstracts and 110 full publications were reviewed. Sixty-six articles met all thepreviously described criteria. The number of articles varied for each of the 5 regions, studies



reporting from the Americas (n = 20), Europe (n = 17), Asia (n = 21), Africa (n = 3) and Ocea-nia (n = 5) (Fig 1). Within the Americas, European and Asian regions, the reports analyseddescribed 8 different countries within each region. In contrast, within the African and Oceaniaregions, reports were restricted to only two countries each; South Africa and Egypt and Austra-lia and New Zealand, respectively.

Clinical variants of OM presentation included in this review are AOM (n = 38 studies;58%), OME/COME (n = 24; 36%), RAOM/AOMTF (n = 9; 14%) and RAOM/OME (n = 1,2%) within the included studies (n = 66). This percentage is greater than 100% due to 4 studies,in which, more than one type of OM was investigated [16–19]. All studies included in thisreview, reported microbiological findings fromMEF collected via clinical perforation of thetympanic membrane.

Fig 1. Literature search and selection.

doi:10.1371/journal.pone.0150949.g001



Microbiology of AOM, RAOM/AOMTF and COMEBacterial isolation overview. This systematic review includes 66 reports of bacterial isola-

tion, using culture techniques, fromMEF samples of patients experiencing AOM (n = 38,mean percentage of samples positive for bacteria 62%; range 25%–95%), RAOM/AOMTF(n = 12; mean percentage of samples positive for bacteria 45%, range 27%- 68%) and OME/COME (n = 24, mean percentage of samples positive for bacteria 36%; range 14%-73%) fromacross the world. Eight studies reported bacterial identification for more than one clinical pre-sentation of OM. Bacterial isolation and identification were performed using PCR and bacterialculture in 17 reports, with only 2 reports from AOM patients [18, 20] rather than RAOM,OME or COME patients.

The wide variety of bacteria isolated in each of the reviewed studies, from patients withAOM, RAOM/AOMTF and OME/COME are shown in S1–S3 Tables, respectively. The fre-quency of detection of S. pneumoniae,H. influenzae andM. catarrhalis in these reports, iden-tify them as the three predominant bacterial pathogens. Other bacteria isolated from the MEFincluded: Streptococcus pyogenes or Group A Streptococcus, Staphylococcus aureus, Pseudomo-nas aeruginosa, Staphylococcus epidermidis, Chlamydia trachomatis, Alloiococcus otitidis, Kleb-siella pneumoniae, Escherichia coli. The variety of microbiota isolated and identified in thesereports clearly varied with the experimental design, aims and methodology of each study andthus precludes direct comparisons of the frequency of individual pathogen detection. Thus,this review is focussed on the relative frequency of the predominant bacterial pathogens.

Previous reports for the different clinical presentations were collated and examined region-ally. Identification of the three predominant bacteria within the MEF for children experiencingAOM, RAOM/AOMTF and OME/COME are presented in S1–S3 Tables, respectively. Thesetables summarise the published reports of bacteria isolated and identified fromMEF using bac-terial culture (S1–S3 Tables).

AOM. The bacterium most frequently identified from the MEF from children diagnosedwith AOM, from previous reports across the world, was S. pneumoniae closely followed byH.influenzae (see Table 1). On average, S. pneumoniae detection was significantly higher than H.influenzae when all previous studies were pooled (P<0.036 2-way ANOVA without replica-tion) and these reports included samples collected between 1979 and 2010.

Regionally, in MEF from AOM patients, the average frequency of S. pneumoniae was higherthan NTHi detection for all regions, although there was no significant difference between aver-age detection frequencies for these bacteria in the US (27.8% NTHi versus 28.6% S. pneumo-niae). In contrast, NTHi was the predominant bacteria, comparing average frequencies ofbacterial detection for all 3 regions that reported RAOM/AOMTF; South America, Europe andOceania. Two studies conducted in Costa Rica showed that S. pneumoniae was predominant inRAOM.

Within each world region, the trend toward more frequent detection of S. pneumoniae, thatis, the average frequency of S. pneumoniae was higher than NTHi detection for all regions.Indeed, where multiple reports were available for the same country within a region, reportsfrom a number of countries including Finland, Colombia, USA and Japan showed changes inthe predominant bacteria identified. Multiple studies from the USA provide evidence of tem-poral trends in the detection frequency changing from S. pneumoniae in studies recruitingbetween 1989–1998 [21–23] to H. influenzae for similar studies recruiting between 2005–2009[24, 25], then back to S. pneumoniae 2006–2010 [26] (Table 1).

For children experiencing AOM (Table 1), 29/38 of studies reported S. pneumoniae as thepredominant bacteria compared toH. influenzae (primarily non-typeable) being clearly pre-dominant bacteria in 6/38 studies (equivalent detection in 3 reports). The overall average



Table 1. S. pneumoniae,H. influenzae andM. catarrhalis detected fromMEF of patients with AOM.

Countries/regions OM Agea Size Bacteriab Ref.

Hi Spn Mcat

South America

Argentina (1990–1992) AOM 1m—11y 161 13.7% 20.5% 1.2% [27]

Argentina (1996–1997) AOM 15d—24m 367 24.5% 27.8% 0.5% [28]

Brazil (1990–1995) AOM 2m—5y 300 7.0% 16.0% 5.0% [29]

Chile (1998–1999) AOM 3m—9y 170 24.1% 37.1% 1.2% [30]

Chile (1998–2002) AOM 3m—9y 543 28.7% 39.8% 4.2% [31]

Costa Rica (1992–1997) AOM 4m—12y 398 14.1% 29.6% 1.5% [32]

Costa Rica (1999–2001) AOM 4m—12y 102 13.7% 42.2% 2.9% [16]

Costa Rica (2002–2007) AOM 2m—8y 880 23.0% 27.7% 8.8% [17]

Colombia (1979–1985) AOM 18d—11y 111 35.1% 23.4% 0.9% [33]

Colombia (2008–2009) AOM 3m—5y 83 28.9% 30.1% NDc [34]

Mexico (2008–2009) AOM 3m—5y 99 31.3% 31.3% 2.0% [35]

Venezuela (2008–2009) AOM 3m—5y 82 41.5% 23.2% 1.2% [36]

Average 23.8% 29.1% 2.7%

Max 41.5% 42.2% 8.8%

Min 7.0% 16.0% 0.5%

North America

The US (1989–1993) AOM 2m—7y 815 23.3% 24.9% 15.0% [21]

The US (1993–1995) AOM � 6y (86%) 159 28.3% 38.4% 25.8% [22]

The US (1989–1998) AOM 2m—7y 982 23.3% 25.8% 13.1% [23]

The US (2005–2009) AOM 2m—3y 184 31.5% 30.4% 7.6% [24]

The US (2006–2008) AOM 6m—3y 170 31.8% 20.6% 7.6% [25]

The US (2008–2010) AOM 4m—3y 208 28.4% 31.7% 13.9% [26]

Average 27.8% 28.6% 13.8%

Max 31.8% 38.4% 25.8%

Min 23.3% 20.6% 7.6%

Europe

Spain (1989–1995) AOM 1m—14y 104 28.8% 34.6% 1.0% [37]

Finland (1980–1985) AOM � 3m 85 11.8% 18.8% 3.5% [38]

Finland (1990–1992) AOM 3m—8y 118 18.6% 22.0% 10.2% [20]

Finland (1994–1995) AOM 2m—2y 772 22.5% 26.0% 22.9% [39]

Finland (1998–1999) AOM 7m—6y 79 29.1% 49.4% 27.8% [40]

Germany (2008–2010) AOM 3m—5y 24 29.1% ND 4.2% [41]

Average 23.3% 30.2% 11.6%

Max 29.1% 49.4% 27.8%

Min 11.8% 18.8% 1.0%

Asia

Israel (1995–1996) AOM 3m—3y 249 45.4% 37.8% 4.8% [42]

Israel (1995–1999) AOM <2m 137 29.9% 40.9% 2.2% [43]

Israel (1996–2003) AOM 3m—3y 771 54.6% 35.9% 1.2% [44]

Japan (1979–1980) AOM � 16y 406 32.0% 49.9% ND [45]

Japan (2001–2002) AOM 1m—9y 81 7.4% 9.9% 7.4% [46]

Japan (2003) AOM Children 138 21.7% 21.0% 3.6% [47]

Japan (2002–2004) AOM � 10y 1092 20.8% 22.3% 4.4% [48]

Japan (2006) AOM 9m—8y 40 5.0% 12.5% ND [18]

Taiwan (2004) AOM 4m—13y 96 13.5% 16.7% ND [49]

(Continued)



Table 1. (Continued)

Countries/regions OM Agea Size Bacteriab Ref.

Hi Spn Mcat

Thailand (2008–2009) AOM 3m—5y 107 17.8% 24.3% 6.5% [50]

Turkey (1998–2000) AOM 6m—10y 78 14.1% 23.1% 5.1% [51]

Turkey (2002–2004) AOM 6m—12y 180 13.3% 25.6% 10.0% [52]

Turkey (2003–2004) AOM 6m—12y 120 13.3% 23.3% 8.3% [53]

Average 22.2% 26.4% 5.4%

Max 54.6% 49.9% 10.0%

Min 5.0% 9.9% 1.2%

Africa

South Africa (1999) AOM 2m—7y 173 5.2% 20.20% 1.20% [54]

Average (all regions) 23.1% 27.8% 7.0%

Max (all regions) 54.6% 49.9% 27.8%

Min (all regions) 5.0% 9.9% 0.5%

a: d = day; m = month; y = year;b: Hi: H. influenzae; Spn: S. pneumoniae; Mcat: M. catarrhalis;c ND: Not detected

doi:10.1371/journal.pone.0150949.t001

Table 2. S. pneumoniae,H. influenzae andM. catarrhalis detected fromMEF of patients with RAOM/AOMTF.

Countries OM Agea Size Bacteriab Ref.

Hi Spn Mcat

South America

Costa Rica (1999–2001) RAOM 4m—11y 98 12.2% 32.7% NDc [16]

Costa Rica (1999–2001) AOMTF 4m—11y 76 19.7% 17.1% ND [16]

Costa Rica (2002–2007) RAOM 2m—8y 138 26.1% 29.0% 3.6% [17]

Costa Rica (2002–2007) AOMTF 2m—8y 90 24.4% 21.1% 4.4% [17]

Europe

France (2007–2009) RAOM/ AOMTF 3m—3y 143d 31.5% 31.5% 1.4% [55]

Netherlands (2008–2009) RAOM 7m—6y 25 24.0% ND 4.0% [19]

Spain (2008–2010) RAOM/ AOMTF 3m—3y 77 41.6% 16.9% 1.3% [13]

Oceania

Australia (2007–2009) RAOM 7m—3y 143 13.3% 5.6% 6.3% [56]

Australia (2007–2009) RAOM 9m—3y 38 15.8% 7.9% ND [57]

New Zealand (2011) RAOM/OME 3m—3y 325e 19.4% 8.0% 8.0% [58]


Max (all regions) 41.6% 32.7% 8.0%

Min (all regions) 12.2% 5.6% 1.3%

a d = day; m = month; y = year;b Hi: H. influenzae; Spn: S. pneumoniae; Mcat: M. catarrhalis;c ND: Not detected;d 18% from spontaneous otorrhoea;e More than 60% of patients had RAOM




frequency of detection for all studies was 27.8% (range 9.9%-49.9%) for S. pneumoniae com-pared to 23.1% (range 5.0%-54.6%) for H. influenzae,M. catarrhalis was detected in 34/38studies at an average frequency of detection of 7.0% (range 0.5%–27.8%). Although detectedless frequently overall,M. catarrhalis frequency, does not demonstrate a consistent temporaltrend toward increasing detection in reports of AOM. Individual countries such as Finland,Costa Rica and Chile exhibit a temporal trend toward increasing rates of identification. In con-trast, reports from the USA, Japan and Turkey show no consistent trend in identificationswhilst reports from Israel do not support any temporal change inM. catarrhalis detection overtime. Previous reports investigated chronic or longer term OM presentations such as RAOM/AOMTF (Table 2) or OME/COME (Table 3).

Examination of the RAOM/AOMTF reports (n = 7 and n = 3, respectively) did not demon-strate a predominant bacteria overall (Table 2) although the average bacterial detection fre-quency for H. influenzae was demonstrated as 22.8% (range 12.2%-41.6%) in comparison to18.6%; (range 5.6%-32.7%) for S. pneumoniae.M. catarrhalis detection was very low (notdetected in 3 of 10 reports) and average detection frequency from bacterial culture was 4.1%(range 1.3%-8%). There were no significant regional or temporal trends observed, due to thepaucity of studies and the challenge arising from recruitment of overlapping clinical presenta-tions (RAOM/AOMTF) within different study designs.

OME/COME. Overall, MEF samples from patients diagnosed with OME/COME were lesslikely to be culture positive for the 3 predominant bacteria in comparison to MEF samplesfrom patients diagnosed with AOM. This is evidenced by the average frequency of detectionfor H. influenzae and S. pneumoniae from OME/COME patients being approximately half thatreported from AOM patients (11.6% vs 23% and 6.5% vs 27%) for these bacteria respectively.

Globally,H. influenzae was the predominant bacteria identified within the MEF of patientsexperiencing OME/COME (P<0.001 2-way ANOVA without replication) with the averagedetection frequency was 11.6% (range 3.2%-21.6%) compared to S. pneumoniae detection 6.5%(range 1.3%-16.2%).

Regionally, H. influenzae was identified most frequently from patients with OME/COME inSouth and North America, Europe, Asia, Africa and Oceania however in the Asian and Euro-pean regions, four of the nine (44%) and two of nine (22%) of studies reported that S. pneumo-niae was predominant, respectively.

Within this review, the bacterial pathogens associated with OM in the MEF were examinedregionally and evidence of development of antibiotic resistance recorded. Where resistance wasreported, it was included in the summary for each region provided below.

Regional variation of H. influenzae strains and Antibiotic sensitivityOverall, there is a general paucity of reports in whichH. influenzae strains and antibiotic sensi-tivity have been reported from the MEF of OM patients. In general, the available reports high-light the high proportion of NTHi isolates compared to other H. influenzae strains identifiedfrom AOM patients across almost all world regions; the Americas, Europe, Asia and Oceania,with no evidence available from Africa. The proportion ofH. influenzae isolates from AOMpatients that were β-lactamase positive ranged less than 20% for most regions to over 83% in areport fromMexico [35] with no consistent patterns recognisable in regions where multiplereports were available. H. influenzae isolates from COME patients generally reported a propor-tion of β-lactamase positive strains. The majority of reports from across the world identifiedvery high proportions ofM. catarrhalis strains as β-lactamase positive.

In South America, AOM-associatedH. influenzae strains isolated within South America arepredominantly NTHi, ranging from 63% [36] to 100% [17, 35] and H. influenzae types a, b, c,



Table 3. S. pneumoniae,H. influenzae andM. catarrhalis detected fromMEF of patients with OME/COME.

Countries/regions OM Age Size Bacteria* Ref.

Hi Spn Mcat

Americas

Brazil (2001–2002) OME/COME 11m—10y 128 10.2% 6.2% 3.9% [59]

The US (1995) COME 9m—15y 97 21.6% 5.2% 5.2% [60]

Average 15.9% 5.7% 4.6%

Max 21.6% 6.2% 5.2%

Min 10.2% 5.2% 3.9%

Europe

Finland (1981) COME 5m—15y 110 3.6% 2.7% 0.9% [61]

Wales (1986–1987) COME � 7y (67%) 259 12.4% 2.7% 0.4% [62]

England (1989) COME < 10y (Mostly) 102 16.7% 3.9% 1.0% [63]

Finland (1993–1994) OME 5m—12y 165 8.5% 10.3% 6.7% [64]

Finland (1993–1994) OME/COME < 12y 123 14.6% 11.4% 6.5% [65]

Finland (1996–1997) OME 1y—9y 67 9.0% 3.0% 9.0% [66]

Spain (2007) OME/COME 1y—12y 40 12.5% 2.5% ND** [67]

Netherlands (2008–2009) COME < 6y 94 19.1% 5.3% 9.6% [19]

UK (2012) COME � 18y (83%) 62 3.2% 6.50% 4.8% [68]

Average 11.1% 5.4% 4.9%

Max 19.1% 11.4% 9.6%

Min 3.2% 2.5% 0.4%

Asia

Turkey (1999) COME 2y—14y 37 10.8% 16.2% 2.7% [69]

Korea (2004–2008) COME 2y—7y 289 ND 1.7% ND [70]

Korea (2000–2002) COME 1y—11y 278 7.9% 1.4% ND [71]

Japan (2006) COME 9m—8y 76 5.3% 1.3% 1.3% [18]

Japan (1988) OME � 15y (73%) 613 5.7% 3.3% 0.5% [72]

Iran (2007–2008) COME 2y—13y 63 9.5% 15.9% 9.5% [73]

Iran (2009–2010) COME 1y—12y 63 4.8% 9.5% 9.5% [74]

Lebanon (1996–1997) OME 2y—10y 47 19.1% ND 4.3% [75]

Lebanon (2009–2010) COME < 13y 107 19.6% 8.4% 3.7% [76]

Average 10.3% 7.2% 4.5%

Max 19.6% 16.2% 9.5%

Min 4.8% 1.3% 0.5%

Africa

Egypt (1993) COME 15m—8y 104 18.3% 5.8% 8.7% [77]

Egypt (2003–2008) OME/COME 3y—10y 50 ND 12.0% 14.0% [78]

Average 18.3% 8.9% 11.4%

Max 18.3% 12.0% 14.0%

Min 18.3% 5.8% 8.7%

Oceania

Australia (1995–2000) COME 11m—10y 45 6.7% ND 4.4% [79]

New Zealand (1994) COME 11m—8y 105 16.2% 8.6% 12.4% [80]

Average 11.5% 8.6% 8.4%

Max 16.2% 8.6% 12.4%

Min 6.7% 8.6% 4.4%


(Continued)



d and f which were identified in AOM patients in Chile [30] and Venezuela [36]. Less than20% of all AOM-associated H. influenzae strains were β-lactamase positive [17, 28–30, 34, 36].In contrast, a recent study fromMexico reported that 83% of AOM-associated H. influenzaestrains were β-lactamase positive [35], whilst other reports demonstrated that 100% of AOM-associatedM. catarrhalis strains were β-lactamase positive [17, 29].

In North America, all AOM-relatedH. influenzae strains reported were non-typeable [81],with 20%-50% of these strains being β-lactamase positive [24, 81]. Furthermore, 100% ofM.catarrhalis strains isolated from AOM patients were β-lactamase positive [81, 82]. Recently,two studies have reported thatH. influenzae was the bacteria detected most often in MEF fromchildren with AOM since the introduction of pneumococcal vaccine into the NIP [24, 25].

In Europe, a recent study from Germany, reported 86% of AOM-relatedH. influenzaestrains were non-typeable, with H. influenzae type b and f also being identified in low frequen-cies (<10%) [41]. In France, more than 15% of AOM-associated H. influenzae strains isolatedwere β-lactamase positive [14, 55] and 95% of AOM-relatedM. catarrhalis strains were β-lacta-mase positive [14].

β-lactamase positive strains ofH. influenzae were reported for 12% and 40% of strains iso-lated from OME/COME patients in England [63] and Spain [67], respectively.

In Asia, more than 80% ofH. influenzae strains associated with AOM in this region werenon-typeable andH. influenzae type a and b were identified in low frequencies (<20%) [52, 53,83, 84], except two studies in Thailand [50] and Taiwan [49] where type b was reported in 63%and 100% of H. influenzae strains. Less than 20% of isolatedH. influenzae strains were β-lacta-mase positive [43, 52, 53, 83, 85]. In contrast, all OM-associatedM. catarrhalis strains isolatedin Israel [43, 83, 85] and over 50% of strains isolated in Turkey [52, 53] were β-lactamase posi-tive. Furthermore, more than 50% of H. influenzae isolates from children with COME in Leba-non were β-lactamase-negative ampicillin-resistant strains [76], however these strains were nottyped.

In Africa, there are no reports of H. influenzae type from AOM patients in the Africanregion, however a single report from Egypt confirmed that more than 80% and 60% of COME-associated H. influenzae andM. catarrhalis strains isolated were β-lactamase positive, respec-tively [77].

In Oceania, In Oceania, H. influenzae isolates from RAOM patients in Australia were allnon-typeable strains with 17% of these strains were β-lactamase positive [56]. Similarly, inNew Zealand, 95% of OM-related H. influenzae strains were non-typeable [58] and a singlereport of COME-associatedH. influenzae strains in New Zealand reported that 6% (1/17) weretype b and 94% (16/17) were not type b [80].

Table 3. (Continued)

Countries/regions OM Age Size Bacteria* Ref.

Hi Spn Mcat

Max (all regions) 21.6% 16.2% 14.0%

Min (all regions) 3.2% 1.3% 0.4%

a d = day; m = month; y = year;b Hi: H. influenzae; Spn: S. pneumoniae; Mcat: M. catarrhalis;c ND: Not detected




Changing microbial pattern pre- and post-PCV7 introductionPCV7 has been introduced into NIPs since the year 2000. For a number of regions, includingSouth America, Asia and Africa, this vaccine has either not yet been introduced to the NIP oronly been recently introduced. For example, PCV7 was introduced into Israel’s NIP and SouthAfrica’s NIP in 2009 [86, 87] or was recommended for high risk groups in Mexico in 2006 andsubsequently introduced in to Mexico’s NIP in 2008 [35]. Furthermore, the introduction ofPCV7 into NIP or high risk groups within individual countries in a region may also impact theaetiology of OM within that region.

The demonstrated impacts of PCV7 immunisation on the aetiology of OM, since its intro-duction in 2000 includes: reduction of PCV7 serotype detection in OM, replacement of PCV7serotypes in AOM by non-vaccine serotypes, particularly serotypes 19A and 3 and changes indominant otopathogens between H. influenzae and S. pneumoniae.

In South America, PCV7 was introduced partially in Costa Rica and Venezuela in 2004 [17,36, 88] and Mexico in 2006 [35]. The vaccine was implemented in NIP of Costa Rica in 2009[89] and Mexico in 2008 [35, 89].

The overall rates of pneumococcal OM were reduced in the South American region afterPCV7 introduction, as did the ratio of S. pneumoniae and H. influenzae isolated from patientswith OM. For example, in Costa Rica, the frequency of S. pneumoniae isolated from patientswith OM decreased from 42% (1999–2001) to 28% (2002–2007) after PCV7 introduction in2004, while the percentage of isolated H. influenzae increased from 14% (1999–2001) to 23%(2002–2007) [16, 17]. Similarly, in Mexico, the proportion of S. pneumoniae isolated frompatients with OM dropped from 52% (2002–2003) to 31% (2008–2009) [35, 90].

Persistence of PCV7 serotypes was demonstrated in early follow up studies in Costa Ricaand Mexico, within 3–4 years of PCV7 introduction, in which PCV7 serotypes were identifiedin 60% and 40% of pneumococcal isolates from patients with OM respectively in these coun-tries [35, 91].

Emergence of non-vaccine serotypes was observed in both countries, with serotype 3accounting for 10% of identified pneumococcal isolates in Costa Rica [88, 92] and serotypes19A, 15B and 28A identified in 23%, 10% and 7% of serotypes from OM patients in Mexico,after PCV introduction [35]. In the longer term, 4 years after PCV7 introduction in Venezuela,the predominant pneumococcal serotype isolated from children with OM was serotype 19A(41% of identified serotypes), followed by other non-PCV7 serotypes such as 6A (9%), 11 (9%)and 15 (9%) [36]. Overall, the emergence of non-PCV7 serotypes was concurrent with an over-all reduction in the frequency of pneumococcal isolation from children with OM.

In North America, PCV7 was first introduced in the United States to prevent invasive pneu-mococcal disease in 2000 and 93.6% of children aged from 19 to 35 months had received atleast 3 doses of PCV7 [93].

After PCV7 introduction, the predominant pathogen identified in OM in the US changedfrom S. pneumoniae toH. influenzae. Prior to PCV7 introduction, S. pneumoniae was the mostcommon bacteria isolated from patients with AOM [22, 23, 82] but 1–3 years following PCVimplementation,H. influenzae emerged as the most common isolates from patients with OM[12, 24].

Changes in the S. pneumoniae serotypes causing AOM had been observed after PCV7 intro-duction in the United States. Before PCV introduction, vaccine serotypes accounted for morethan 70% of pneumococcal strains isolated from middle ear samples of children with AOM[94]. One to two years post-vaccine introduction, these serotypes accounted for about 50% ofOM-associated pneumococcal strains isolated from middle ear samples [94] and six to eightyears later they were not observed/reported amongst pneumococcal strains isolated from MEF



samples of children with OM [12]. Serotype 19A was the predominant serotype and accountedfor 35% of OM-related S. pneumoniae strains isolated MEF samples [12]. Other non-PCV7serotypes of S. pneumoniae, such as 3, 6A, 6C, 15, 23B and 11, were identified with lower fre-quencies in MEF samples of OM children [12, 94]. In recent years, although the pneumococcalserotypes are non-PCV7, frequencies of S. pneumoniae and H. influenzae isolates are equiva-lent [12].

In Europe, a 2008–2009 survey of 32 countries reported that only 17 countries recom-mended universal PCV7 immunization in children. The remaining countries had only used thevaccine for groups at high risk of infection or not used the vaccine universally [95]. Impacts ofthe vaccine on OM aetiology and pneumococcal serotypes identified from OM patients wereobserved in countries where the vaccine was introduced.

Following PCV7 introduction, the dominant otopathogen changed from S. pneumoniae toH. influenzae in France and Greece. In France, prior to the introduction of PCV7, S. pneumo-niae was the otopathogen most commonly identified accounting for 54% of isolates and fol-lowed byH. influenzae (34%), however 1–2 years following PCV7 introduction, H. influenzaebecame the dominant bacteria, identified within approximately 46% of isolates and followed byS. pneumoniae (45%) [14]. This change in pattern was evident until 3–4 years post PCV7implementation, where S. pneumoniae was again the most commonly isolated specie withmostly non-PCV7 serotypes [14]. Similarly, in Greece, S. pneumoniae isolation from the mid-dle ear samples of children with AOM decreased from 42% to 31% post PCV introduction,while H. influenzae increased slightly from 34% to 37% [96].

Introduction of the PCV7 vaccine resulted in changes in the pneumococcal serotypes iso-lated from patients with OM. In France, pre-PCV7 vaccine introduction (this period, prior to2004, is identified by the criterion that the rate of complete vaccination in infants under 2 yearsof age was below 50%), PCV7 serotypes accounted for more than 50% of S. pneumoniae associ-ated with OM but 1–4 years post-PCV7 introduction, this decreased to less than 30% [14]. Sim-ilarly, in Spain, the percentage of PCV7 serotypes dropped from more than 60% to below 10%of pneumococcal strains isolated from patients with OM [97–99]. Following PCV7 introduc-tion, the frequency of the non-PCV7 serotype 19A, increased from 20% to 50% in patients withOM in France [14] and from 8% to 35% in Spain [98, 99].

In Asia, Singapore and Israel introduced PCV7 into their NIPs in 2009 [86, 100–102]. Stud-ies from Israel reported that PCV7 serotypes accounted for approximately 50% of S. pneumo-niae strains associated with OM prior to the vaccine introduction [102, 103]. No studies havebeen identified in this review, which report rates following PCV7 introduction on microbialaetiology of OM or from other countries in this region.

In Africa, PCV7 was introduced into the NIP of South Africa in 2009 [87] and Gambia andRwanda in 2010 [104]. A single study from South Africa examined S. pneumoniae serotypespre- PCV7 and showed that PCV7 serotypes accounted for more than 90% of OM-associatedS. pneumoniae [54].

In Nigeria, a shift in the predominant bacteria amongst the three main otopathogensdetected fromMEF samples of patients with AOM was observed, although information onPCV7 introduction has not been reported. In the early 2000s, S. pneumoniae was reported asthe predominant otopathogen, followed byH. influenzae andM. catarrhalis [105], howevermore recently, H. influenzae was reported as the predominant otopathogen, followed byM cat-arrhalis and S. pneumoniae [106].

In Oceania, PCV7 was licenced and recommended for Australian Aboriginal childrenunder 2 years of age and other children at high risk of pneumococcal infection in 2001. Thisrecommendation was extended to all children in Australia in 2005 [107]. Pneumococcal sero-types recovered from middle ear samples showed progressive change after PCV7



immunisation. In 2000–2003, pre-PCV7 introduction, vaccine serotypes accounted for approx-imately 70% of the pneumococci isolated from ear discharges (ear swabs) of Australian chil-dren below 15 years of age [108]. In 2007–2009, 2–4 years post-PCV7 inclusion in NIP, non-PCV7 serotypes replaced the vaccine serotypes being isolated from middle ear samples fromnon-Aboriginal children with OM, accounting for more than 80% of identified serotype, ofthese serotype 19A made up 40% [56].

In contrast to other regions, H. influenzae has always been reported as the most commonotopathogen isolated from both Aboriginal and non-Aboriginal children with OM in Australia[56, 79, 109–111]. Thus, there has not been a shift between H. influenzae and S. pneumoniae asthe predominant otopathogen observed pre- and post- PCV7 introduction in the country. Sim-ilar observations are reported for New Zealand, with H. influenzae identified as the most com-mon bacteria isolated from children with OM (RAOM or OME/COME) before and afterpneumococcal vaccine introduction (2008)[58, 80, 112].

Improvement of detection of existing and new otopathogens by PCRThe frequency of bacterial detection within MEF samples from children with OM have beencompared using both culture and PCR. Across the world, 17 studies have reported bacterialidentification, the majority (n = 15) examined MEF from patients with COME or RAOM, with2 reports from AOM patients.

Overall, PCR detection methods improved the detection frequency of bacteria although themagnitude of the improvement varied between individual reports and ranged from no differ-ence to 9-fold increase in bacterial detection. On average, the improvement in the frequency ofdetection for bacteria identified within the MEF of children with OM is 3.2 times higher usingPCR compared to bacterial culture, for both S. pneumoniae and H. influenzae. The sensitivityof PCR techniques also improvedM. catarrhalis detection by 4.5- fold, on average across allstudies (Table 4).

The enhanced sensitivity of detection observed using PCR has resulted in detection ofM.catarrhalis and S. pneumoniae in culture negative samples [18, 19, 71, 113].

PCR confirmation of the same predominant bacterium being identified within the samesample was observed in 10/17 reports however, in the 6 reports showing a different predomi-nant bacteria, 4/6 changed toM. catarrhalis and 2/6 changed toH. influenzae. A predominantbacterium was not identified in one study. Where PCR identified a different predominant bac-terium, there was no recognisable pattern compared to the culture result, with 3 changes fromH. influenzae and 3 from S. pneumoniae as the predominant bacterium from the culture.

Globally, and consistent with the results from bacterial culture, PCR results confirmed over-all that H. influenzae was the most frequently detected bacterium in the MEF of childrenexperiencing COME/RAOM (11/15 studies).M. catarrhalis was predominant in 3/4 studiesand all three bacteria were equally frequent in one study from Iran. The two studies that exam-ined AOM patients MEF, identifiedM. catarrhalismost frequently, although Alloiococcus otiti-dis (A. otitidis) was equally or more frequently identified in these reports.

Regionally, from PCR results, H. influenzae was detected most frequently in the Americas,Asia and Oceania. In Europe, H. influenzae andM. catarrhalis were equally frequent (2/4 stud-ies each).

PCR detection for A. otitidis from the MEF of OM patients, has been reported from a rangeof countries, including Turkey [114], Japan [18], Iran [74], Finland [20, 65, 66], Sweden [115],the United Kingdom [116] and the United States [25, 117]. Collectively, A. otitidis was detectedin 20% to 60% of MEF samples from OM patients by PCR, but was largely undetected using cul-ture (Table 4). Furthermore, A. otitidis was more frequently identified in patients with OME



compared to patients with AOM [18, 117] and, was the most commonly detected bacteria in themiddle ear of patients with OME/COME, detected using PCR [18, 74, 117, 118]. The use of PCRmethodologies has improved detection rates for organisms that are difficult to culture, for exam-ple, detection of A. otitidis within MEF samples from patients with OM requires more than 3days in culture, thus PCR analysis of MEF samples has improved detection of this organism andhas resulted in consideration of A. otitidis having a role in OM pathogenesis [58, 67, 74, 119].

DiscussionThe frequency of bacterial pathogens of OM, identified within the middle ear fluid using pri-marily microbial culture, but also PCR, were reviewed over a range of clinical presentations,using reports published since 1970. This review focussed on pathogens actually identifiedwithin the MEF rather than extrapolation of upper respiratory tract pathogens coincident inthe nasopharynx. Improved understanding of the polymicrobial nature of OM and its variedclinical presentations is informed by the identification of specific bacteria (and viruses) withinthe middle ear.

Overall, the frequency of bacterial isolation fromMEF samples of patients with AOM wasgenerally higher than that from other presentations including RAOM/AOMTF or OME/COME. This finding is consistent with recognition of AOM as an acute, frequently bacterialcaused infection, whilst OME and COME are considered as a consequence of a previous AOM

Table 4. Comparison of otopathogen detection in MEF using culture and PCR.

Countries OM Culture PCR Ref.

Bacteria Spn Hi Mcat Aota Bacteria Spn Hi Mcat Aot

Brazil (2001–2002) COME/ RAOM 20% 6% 10% 4% - 62% 13% 39% 10% - [59]

US (1995) COME 32% 5% 22% 5% - 131% 30% 55% 46% - [60]

Finland (1990–1992) AOM 51% 22% 19% 10% NDb 58% 20% 11% 27% 25% [20]

Finland (1993–1994) COME 33% 11% 15% 7% ND 131% 35% 33% 63% 20% [65]

Finland (1996–1997) OME 21% 3% 9% 9% ND 76% 21% 18% 37% 46% [66]

Netherlands (2008–2009) RAOM 28% ND 24% 4% - 84% 13% 42% 29% - [19]

Netherlands (2008–2009) COME 34% 5% 19% 10% - 63% 7% 32% 24% - [19]

Iran (2007–2008) COME 36% 16% 10% 10% - 151% 19% 95% 37% - [73]

Iran (2009–2010) COME 25% 10% 5% 10% 24% 32% 11% 11% 10% 40% [74]

Korea (2000–2002) COME 9% 1% 8% ND - 45% 5% 29% 11% - [71]

Turkey (1999) COME 30% 16% 11% 3% - 143% 57% 70% 16% - [69]

Japan (2005) COME 10% ND 5% 5% - 48% 12% 23% 13% - [113]

Japan (2006) OME 7% 1% 5% 1% ND 27% 8% 12% 7% 61% [18]

Japan (2006) AOM 18% 13% 5% ND ND 41% 13% 8% 20% 50% [18]

Australia (2007–2009) RAOM 25% 6% 13% 6% - 66% 6% 47% 13% - [56]

Egypt (2003–2008) OME/ COME 26% 12% - 14% - 92% 48% - 56% - [78]

New Zealand (2011) RAOM/ OME 19% 8% 19% 8% - 65% 23% 43% 39% - [58]

PCR vs culture average 3.3

Hi (PCR vs culture) average 3.2

Spn (PCR vs culture) average 3.2

Mcat (PCR vs culture) average 4.5

a Aot: A. otitidis;b ND: Not detected




episode, or inflammation resulting from the presence of bacterial biofilm or intracellular infec-tion within the middle ear epithelial cell [57]. COME and OME may also result from inflam-matory and allergic conditions such as rhino-sinusitis [120].

This review confirms OM as a polymicrobial disease, with the same predominant microbesidentified in MEF samples from patients with AOM, RAOM, AOMTF, OME and COME glob-ally. Consistent with previous reports, the three bacteria most frequently identified in the MEFfrom culture and considered causal for OM were S. pneumoniae, H. influenzae andM. catar-rhalis. In Latin American countries, Streptococcus pyogenes (S. pyogenes) was also frequentlyidentified as a predominant bacterium, pathogenic for OM. Increasingly, A. otitidis has beenidentified within the MEF using PCR techniques, although this bacterium’s role in OM patho-genesis is still under investigation [121, 122]. Culture based detection of A. otitidis within MEFsamples from patients with OM requires more than 3 days in culture and thus this organismmay not have been successfully identified and reported in previous studies [58, 67, 74, 119].

S. pneumoniae remains the most frequently detected bacteria from patients with AOM inAsia, Africa, America and Europe. No significant differences between the clinical manifesta-tions of AOM, resulting from the different pneumococcal serotypes, were identified but in gen-eral, PCV7 serotypes are reported to cause more otalgia or ear ache than non-PCV7-serotypes[123]. Additionally, children with AOM caused by S. pneumoniae are reported to experiencehigher fever and more redness of tympanic membrane than those with AOM caused byH.influenzae orM. catarrhalis [124], thus increasing the likelihood of presentation for medicaltreatment. AOM associated with S. pneumoniae is less likely to resolve without treatment whencompared to that caused byM. catarrhalis [125].

Overall, use of PCV7 has initially reduced the incidence of AOM caused by S. pneumoniaeand altered the serotypes causal for AOM. Replacement of PCV7 serotypes by non-vaccineserotypes has been observed in MEF samples of children with OM in all countries where thevaccine has been used, with serotype 19A, a predominant serotype recovered from vaccinatedchildren with OM. A new genotype of this serotype, isolated from children with AOM wasresistant to all USA Food and Drug Administration approved antimicrobial drugs for AOMtreatment in children [126]. Therefore, 13-valent pneumococcal conjugate vaccine (PCV13),which includes serotype 19A, may help to reduce the incidence of OM. Since 2010, PCV13 hasbeen licensed and recommended for children in a number of developed countries such as theUnited States, the United Kingdom, Singapore and Australia [101, 127–129].

In contrast to AOM, H. influenzae is most frequently detected bacteria in the MEF ofpatients with RAOM or OME/COME globally and, with the exception of Australia, is the sec-ond most common pathogen in patients with AOM across the world. In Australia, H. influen-zae is the predominant bacterium for AOM and OME for indigenous and non-indigenouschildren [110, 111, 130]. Bilateral AOM, eye symptoms, previous treatment with antibiotics,protracted and recurrent disease, are more likely for AOM caused by H. influenzae when com-pared with caused by other pathogens [131, 132].

In most studies included in this review, less than 50% of H. influenzae strains recoveredfrom patients with OM were β-lactamase producers. β-lactamase is responsible for bacterialresistance to the β-lactam antibiotics, commonly used as the first line of treatment for OM,such as aminopenicillins, cephalosporins, cephamycins and carbapenems [133, 134]. Wide-spread use of these antibiotics for the treatment of OM has contributed to the development ofantimicrobial resistance. β-lactamase-negative ampicillin-resistant (BLNAR) H. influenzaestrains, have been isolated from the MEF of patients with OM in Mexico [35], Lebanon [76]and Japan [135]. The BLNAR strains identified fromMEF in the latter two reports [76, 135]have been isolated elsewhere in the upper respiratory tract and reflect the most frequentlydetected mechanism of ampicillin resistance forH. influenzae across the world [136]. Although



not based on MEF samples, a recent report has described the changing frequency of identifica-tion of BLNAR H. influenzae isolates, both globally and regionally, and confirmed reductionsin detection in Europe and Africa between 2004–2008 and 2009–2012 whilst the proportion ofdetections rose in other regions including Latin America and the Middle East [137].

In this review, NTHi strains were most frequently reported and these strains are unaffectedbyH. influenzae b conjugate vaccination alone (Hib) [138]. Increasingly, inclusion ofH. influen-zae antigens to improve the opportunity for vaccine development is under development [139].

PCV7 introduction has been associated increased identification of H. influenzae in patientswith OM when compared with S. pneumoniae.H. influenzae has gradually replaced S. pneumo-niae as the predominant otopathogen in children with AOM for 1–5 years following PCV7introduction and, in the absence of an approved vaccine for NTHi prevention, this shift couldcreate another challenge for treatment and prevention of OM.

Until recently, most NTHi vaccine research interest was focussed on the 10-valent pneumo-coccal NTHi protein D conjugate vaccine (PHiD-CV10), which contains 10 pneumococcalserotypes and protein D of H. influenzae. This vaccine is immunogenic against protein D andwell tolerated in studies conducted in America, Europe, Asia, Africa [140–145] and Australia[129]. Originally considered a potential candidate to replace PCV7, a recent study from Austra-lia demonstrated that 19% of NTHi strains isolated from the nasal cavity, nasopharynx andbronchoalveolar lavage fluid from asymptomatic carriage or children with bronchiectasis weremissing the hpd genes encoding protein D of NTHi [146]. Further research is needed to trial anumber of already identified NTHi vaccine candidates, with respect to the safety, immunoge-nicity and efficacy against OM [1].M. catarrhalis has been recognised as an otopathogen since1920s with significant increased detection occurring since the 1980’s [147]. Clinically,M. catar-rhalis alone appears to be the least virulent pathogen [148] and is often associated with the firstepisode of disease, a younger age of onset, lower rates of spontaneous tympanic membrane per-foration [149] and increased likelihood of spontaneous resolution [125].M. catarrhalis how-ever, most frequently occurs as a polymicrobial infection [149]. In this review, where culturefindings of predominant bacteria were identified as different to that identified by PCR meth-ods,M. catarrhalis was the new predominant bacterium.Most studies report that more than90% ofM. catarrhalis strains isolated from patients with OM were β-lactamase producers,although lower rates of 40%-50% have been reported from Turkey and Nigeria [52, 53, 106].Whilst several vaccine candidates have been identified forM. catarrhalis, development of a vac-cine strategy forM. catarrhalis is at an early stage [150–152].

Otitis media is associated with other bacteria including S. pyogenes (Group A Streptococ-cus), the fourth ranked otopathogen [82] identified in this review. This bacterium is otopatho-genic and more often identified in older children who experience low rates of fever andrespiratory symptoms but, increased rates of spontaneous perforation and mastoiditis, com-pared with AOM due to other otopathogens [153]. It is speculated that the role of S. pyogenesin OM aetiology maybe under-recognised in some regions.

Increasingly, new organisms such as A. otitidis, Turicella otitidis, Pseudomonas otitidis andCorynebacterium mucifaciens have been detected in patients with OM, but their significance inOM pathogenesis remains unclear [154]. Of these organisms, A. otitidis is the most commonlyidentified in the MEF from patients with OME/COME [18, 117]. A. otitidismay be associatedwith a protracted clinical course, development of the mucoid MEF characteristic of OME [65]and was recently reported to induce inflammation within the middle ear cavity [122, 155] thuspotentially contributing as a secondary pathogen in AOM with perforation [156].

In this review, we have reported on the global detection of bacterial otopathogens in theMEF of children experiencing OM. Although three bacteria, S. pneumoniae, H. influenzae andM. catarrhalis dominate, geographical differences are emerging. These differences relate to



PCV use, antimicrobial treatment and potentially other local regional factors. This highlightsthe need for ongoing surveillance and reporting of the microbiology of OM globally, in orderto better understand the pathogenesis of this disease and to guide development of appropriateintervention strategies.

ConclusionsDespite a paucity of comprehensive prospective studies examining the microbial aetiology ofAOM/RAOM and OME/COME locally within the middle ear of children throughout theworld, we conclude that S. pneumoniae, H. influenzae andM. catarrhalis, have remainedremarkably consistent over the past 40 years, as the predominant bacteria causal for OM.Depending on the region, and the detection method, S. pneumoniae andH. influenzae vie forpredominance as the bacteria most frequently isolated from the middle ear fluid from childrenwith OM. A significant environmental change, the introduction of PCV7 and ongoing moni-toring of serotype replacement throughout many regions of the world, continues to provideevidence of the predominance of these two bacteria in children with OM. Importantly, to iden-tify and address the evolving development of bacterial resistance, such as β-lactamase produc-tion and potential pathogenicity of prevalent microbes such as S. pyogenes and A. otitidisongoing monitoring is essential.

Future research is essential, whether development of vaccines that include multiple oto-pathogens [157], bacterial and viral, or trialling the use of probiotic supplementation to theupper respiratory tract [158] for children at high risk of OM. Investigators should continue toaim to minimise the incidence or indeed prevent OM pathogenesis and its long term sequelaefor vulnerable children across the world.

Supporting InformationS1 Fig. Strategies for searching studies on pathogens of OM in Asia.(DOCX)

S2 Fig. Strategies for searching studies on pathogens of OM in the Americas.(DOCX)

S3 Fig. Strategies for searching studies on pathogens of OM in Africa.(DOCX)

S4 Fig. Strategies for searching studies on pathogens of OM in Europe.(DOCX)

S5 Fig. Strategies for searching studies on pathogens of OM in Oceania.(DOCX)

S1 PRISMA Checklist. PRISMA Checklist.(DOCX)

S1 Table. Proportion of bacteria detected fromMEF samples of patients with AOM.(DOCX)

S2 Table. Proportion of bacteria detected fromMEF samples of patients with RAOM/AOMTF.(DOCX)

S3 Table. Proportion of bacteria detected fromMEF samples of patients with OME/COME.(DOCX)



http://www.plosone.org/article/fetchSingleRepresentation.action?uri=info:doi/10.1371/journal.pone.0150949.s001









AcknowledgmentsAuthors, who, upon email contact, assisted this review by providing clarification or additionalinformation about their research.

Author ContributionsConceived and designed the experiments: AWC HMM. Performed the experiments: CCNHMM. Analyzed the data: CCN HMM RBT AWC. Contributed reagents/materials/analysistools: CCN HMM.Wrote the paper: CCN HMM RBT AWC.

References1. Cripps AW, Kyd J. Bacterial otitis media: current vaccine development strategies. Immunol Cell Biol.

2003; 81(1):46–51. PMID: 12534945

2. Kong K, Coates HLC. Natural history, definitions, risk factors and burden of otitis media. Med J Aust.2009; 191(9):s39–s43. PMID: 19883355

3. Monasta L, Ronfani L, Marchetti F, Montico M, Brumatti LV, Bavcar A, et al. Burden of DiseaseCaused by Otitis Media: Systematic Review and Global Estimates. PLos One. 2012; 7(4):1–12.

4. Rovers MM. The burden of otitis media. Vaccine. 2008; 26(suppl 7):G2–G4. doi: 10.1016/j.vaccine.2008.11.005 PMID: 19094933

5. Physicians AAoF. Otitis media with effusion. Pediatrics. 2004; 113:1412–29. PMID: 15121966

6. Massa HM, Cripps AW, Lehmann D. Otitis media: viruses, bacteria, biofilms and vaccines. Medicaljournal of Australia. 2009; 191(9):s44–s9. PMID: 19883356

7. Chonmaitree T, Ruohola A, Hendley JO. Presence of Viral Nucleic Acids in the Middle Ear Acute OtitisMedia Pathogen or Bystander? Pediatr Infect Dis J. 2012; 31(4):325–30. doi: 10.1097/INF.0b013e318241afe4 PMID: 22173136

8. Weil-Olivier C, van der Linden M, de Schutter I, Dagan R, Mantovani L. Prevention of pneumococcaldiseases in the post-seven valent vaccine era: a European perspective. BMC Infect Dis. 2012; 12(207):1–12.

9. Mokaddas E, Albert MJ. Impact of pneumococcal conjugate vaccines on burden of invasive pneumo-coccal disease and serotype distribution of Streptococcus pneumoniae isolates: an overview fromKuwait. Vaccine. 2012; 30:G27–G40.

10. Isaacman DJ, McIntosh ED, Reinert RR. Burden of invasive pneumococcal disease and serotype dis-tribution among Streptococcus pneumoniae isolates in young children in Europe: impact of the 7-valent pneumococcal conjugate vaccine and considerations for future conjugate vaccines. Int J InfectDis. 2010; 14(3):197–209.

11. Hsu KK, Shea KM, Stevenson AE, Pelton SI. Changing serotypes causing childhood invasive pneu-mococcal disease: Massachusetts, 2001–2007. Pediatr Infect Dis J. 2010; 29(4):289–93. doi: 10.1097/INF.0b013e3181c15471 PMID: 19935447

12. Casey JR, Adlowitz DG, Pichichero ME. New patterns in the otopathogens causing acute otitis mediasix to eight years after introduction of pneumococcal conjugate vaccine. The Pediatric Infectious Dis-ease Journal. 2010; 29(4):304–9. doi: 10.1097/INF.0b013e3181c1bc48 PMID: 19935445

13. Pumarola F, Marès J, Losada I, Minguella I, Moraga F, Tarragó D, et al. Microbiology of bacteria caus-ing recurrent acute otitis media (AOM) and AOM treatment failure in young children in Spain: shiftingpathogens in the post-pneumococcal conjugate vaccination era. Int J Pediatr Otorhinolaryngol 2013;77(8):1231–6. doi: 10.1016/j.ijporl.2013.04.002 PMID: 23746414

14. Dupont D, Mahjuob-Messai F, Francois M, Doit C, Mariani-Kurkdjian P, Bidet P, et al. Evolving micro-biology of complicated acute otitis media before and after introduction of the pneumococcal conjugatevaccine in France. Diagnostic Microbiology and Infectious Disease. 2010; 68:89–92. doi: 10.1016/j.diagmicrobio.2010.04.012 PMID: 20727478

15. Moher D, Liberati A, Tetzlaff J, Altman D, PRISMAGroup. Preferred reporting items for systematicreviews and meta-analyses: the PRISMA statement. PLoS Med. 2009; 6(7).

16. Arguedas A, Dagan R, Soley C, Loaiza C, Knudsen K, Porat N, et al. Microbiology of otitis media inCosta Rican children, 1999 through 2001. Pediatr Infect Dis J. 2003; 22(12):1063–8. PMID:14688566

17. Aguilar L, Alvarado O, Soley C, Abdelnour A, Dagan R, Arguedas A. Microbiology of the middle earfluid in Costa Rican children between 2002 and 2007. Int J Pediatr Otorhinolaryngol 2009; 73:1407–11. doi: 10.1016/j.ijporl.2009.07.005 PMID: 19683349



http://www.ncbi.nlm.nih.gov/pubmed/12534945


http://dx.doi.org/10.1016/j.vaccine.2008.11.005





http://dx.doi.org/10.1097/INF.0b013e318241afe4

http://dx.doi.org/10.1097/INF.0b013e318241afe4


http://dx.doi.org/10.1097/INF.0b013e3181c15471

http://dx.doi.org/10.1097/INF.0b013e3181c15471


http://dx.doi.org/10.1097/INF.0b013e3181c1bc48


http://dx.doi.org/10.1016/j.ijporl.2013.04.002


http://dx.doi.org/10.1016/j.diagmicrobio.2010.04.012

http://dx.doi.org/10.1016/j.diagmicrobio.2010.04.012





18. Harimaya A, Takada R, Hendolin PH, Fujii N, Ylikoski J, Himi T. High incidence of Alloiococcus otitidisin children with otitis media, despite treatment with antibiotics. J Clin Microbiol. 2006; 44(3):946–9.PMID: 16517881

19. Stol K, Verhaegh SJ, Graamans K, Engel JA, Sturm PD, Melchers WJ, et al. Microbial profiling doesnot differentiate between childhood recurrent acute otitis media and chronic otitis media with effusion.Int J Pediatr Otorhinolaryngol 2013; 77(4):488–93. doi: 10.1016/j.ijporl.2012.12.016 PMID: 23369612

20. Leskinen K, Hendolinb P, Virolainen-Julkunen A, Ylikoski J, Jeroa J. Alloiococcus otitidis in acute otitismedia. International Journal of Pediatric Otorhinolaryngology. 2004; 68:51–6. PMID: 14687687

21. Heikkinen T, Thint M, Chonmaitree T. Prevalence of various respiratory viruses in the middle ear dur-ing acute otitis media. Engl J Med. 1999; 340(4):260–4.

22. RodriguezW, Schwartz R, Thorne M. Increasing incidence of penicillin- and ampicillin-resistant mid-dle ear pathogens. Pediatr Infect Dis J. 1995; 14(12):1075–8. PMID: 8745021

23. Patel JA, Nguyen DT, Revai K, Chonmaitree T. Role of respiratory syncytial virus in acute otitismedia: implications for vaccine development. Vaccine. 2007; 25(9):1683–9. PMID: 17156899

24. Grubb MS, Spaugh DC. Microbiology of acute otitis media, Puget Sound region, 2005–2009. ClinicalPediatrics. 2010; 49(8):727–30. doi: 10.1177/0009922810361365 PMID: 20185479

25. Kaur R, Adlowitz DG, Casey JR, Zeng M, Pichichero ME. Simultaneous assay for four bacterial spe-cies including Alloiococcus otitidis using multiplex-PCR in children with culture negative acute otitismedia. Pediatr Infect Dis J. 2010; 29(8):741–5. doi: 10.1097/INF.0b013e3181d9e639 PMID:20335823

26. Casey JR, Kaur R, Friedel VC, Pichichero ME. Acute otitis media otopathogens during 2008 to 2010in Rochester, New York. Pediatr Infect Dis J 2013; 32(8):805–9. doi: 10.1097/INF.0b013e31828d9acc PMID: 23860479

27. MacLoughlin GJF, Barreto DG, Torre Cdl, Pinetta EA, Castillo Fd, Palma L. Cefpodoxime proxetil sus-pension compared with cefaclor suspension for treatment of acute otitis media in paediatric patients.Journal of Antimicrobial Chemotherapy. 1996; 37:565–73. PMID: 9182113

28. Commisso R, Romero-Orellano F, Montanaro PB, Romero-Moroni F, Romreo-Diaz R. Acute otitismedia: bacteriology and bacterial resistance in 205 pediatric patients. International Journal of Pediat-ric Otorhinolaryngology. 2000; 56:23–31. PMID: 11074112

29. Sih TM. Acute otitis media in Brazilian children: analysis of microbiology and antimicrobial susceptibil-ity. The Annals of Otology, Rhinology & Laryngology. 2001; 110(7):662–6.

30. Rosenblut A, Santolaya ME, Gonzalez P, Corbalan V, Avendano LF, Martinez MA, et al. Bacterial andviral etiology of acute otitis media in Chilean children. The Pediatric Infectious Disease Journal. 2001;20(5):501–7. PMID: 11368107

31. Rosenblut A, Santolaya ME, Gonzalez P, Borel C, Cofre J. Penicillin resistance is not extrapolable toamoxicillin resistance in Streptococcus pneumoniae isolated frommiddle ear fluid in children withacute otitis media. The Annals of Otology, Rhinology & Laryngology. 2006; 115(3):186–90.

32. Arguedas A, Loaiza C, Perez A, Vargas F, Herrera M, Rodriguez G, et al. Microbiology of acute otitismedia in Costa Rican children. Pediatr Infect Dis J. 1998; 17(8):680–9. PMID: 9726340

33. Trujillo H, Callejas R, Mejia GI, Castrillon L. Bacteriology of middle ear fluid specimens obtained bytympanocentesis from 111 Colombian children with acute otitis media. Pediatr Infect Dis J. 1989; 8(6):361–3. PMID: 2787494

34. Sierra A, Lopes P, Zapata MA, Vanegas B, Castrejon MM, DeAntonio R, et al. Non-typeable Haemo-philus influenzae and Streptococcus pneumonia as primary causes of acute otitis media in colombianchildren: a prospective study. BMC Infectious Diseases. 2011; 11(4):1–11.

35. Parra MM, Aguilar GM, Echaniz-Aviles G, Rionda RG, Estrada MdLA, Cervantes Y, et al. Bacterial eti-ology and serotypes of acute otitis media in Mexican children. Vaccine. 2011; 29:5544–9. doi: 10.1016/j.vaccine.2011.04.128 PMID: 21596081

36. Naranjo L, Suarez JA, DeAntonio R, Sanchez F, Calvo A, Spadola E, et al. Non-capsulated and cap-sulated Haemophilus influenzae in children with acute otitis media in Venezuela: a prospective epide-miological study. BMC Infectious Diseases. 2012; 12(40):1–10.

37. del Castillo F, Garcia-Perea A, Baquero-Artigao F. Bacteriology of acute otitis media in Spain: a pro-spective study based on tympanocentesis. Pediatr Infect Dis J. 1996; 15(6):541–3.

38. Karma PH, Pukander JS, Sipilä MM, Vesikari TH, Grönroos PW. Middle ear fluid bacteriology of acuteotitis media in neonates and very young infants. Int J Pediatr Otorhinolaryngol. 1987; 14(2–3):141–50.PMID: 3125118

39. Kilpi T, Herva E, Kaijalainen T, Syrjanen R, Takala AK. Bacteriology of acute otitis media in a cohort ofFinnish children followed for the first two years of life. Pediatr Infect Dis J. 2001; 20:654–62. PMID:11465836









http://dx.doi.org/10.1177/0009922810361365


http://dx.doi.org/10.1097/INF.0b013e3181d9e639


http://dx.doi.org/10.1097/INF.0b013e31828d9acc

http://dx.doi.org/10.1097/INF.0b013e31828d9acc












40. Ruohola A, Meurman O, Nikkari S, Skottman T, Salmi A, Waris M, et al. Microbiology of acute otitismedia in children with tympanostomy tubes: prevalences of bacteria and viruses. Clin Infect Dis 2006;43:1417–22. PMID: 17083014

41. Grevers G, Wiedemann S, Bohn J, Blasius R, Harder T, Kroeniger W, et al. Identification and charac-terization of the bacterial etiology of clinically problematic acute otitis media after tympanocentesis orspontaneous otorrhea in German children. BMC Infect Dis. 2012; 12(1):312.

42. Leibovitz E, Piglansky L, Raiz S, Greenberg D, Yagupsky P, Press J, et al. Bacteriologic efficacy of athree-day intramuscular ceftriaxone regimen in nonresponsive acute otitis media. Pediatr Infect Dis J.1998; 17(12):1126–31. PMID: 9877360

43. Turner D, Leibovitz E, Aran A, Piglansky L, Raiz S, Leiberman A, et al. Acute otitis media in infantsyounger than two months of age: microbiology, clinical presentation and therapeutic approach.Pediatr Infect Dis J. 2002; 21:669–74. PMID: 12237601

44. Polachek A, Greenberg D, Lavi-Givon N, Broides A, Leiberman A, Dagan R, et al. Relationshipamong peripheral leukocyte counts, etiologic agents and clinical manifestations in acute otitis media.Pediatr Infect Dis J. 2004; 23(5):406–13. PMID: 15131462

45. Sugita R, Kawamura S, Ichikawa G, Goto S, Fujimaki Y. Studies of anaerobic bacteria in chronic otitismedia. Laryngoscope. 1981; 91(5):816–21. PMID: 7231031

46. Sagai S, Suetake M, Yano H, Yoshida M, Ohyama K, Endo H, et al. Relationship between respiratorysyncytial virus infection and acute otitis media in children. Auris Nasus Larynx 2004; 31(4):341–5.PMID: 15571905

47. Suzuki A, Watanabe O, Okamoto M, Endo H, Yano H, Suetake M, et al. Detection of Human Metap-neumovirus From ChildrenWith Acute Otitis Media. The Pediatric Infectious Disease Journal. 2005;24(7):665–7.

48. Yano H, Okitsu N, Hori T, Watanabe O, Kisu T, Hatagishi E, et al. Detection of respiratory viruses innasopharyngeal secretions and middle ear fluid from children with acute otitis media. Acta Oto-Laryn-gologica. 2009; 129:19–24. doi: 10.1080/00016480802032777 PMID: 18607974

49. Tseng T-C, Chen L-C, Wang P-C, Huang C-H, Chen Y-H. Clinical report on the bacteriology of acuteotitis media in childhood. FJJM. 2007; 5(1):1–8.

50. Intakorn P, Sonsuwan N, Noknu S, Moungthong G, Pircon J-Y, DykeMKV, et al. Haemophilus influen-zae type b as an important cause of culture-positive acute otitis media in young children in Thailand: atympanocentesis-based, multi-center, cross-sectional study. BMC Pediatrics. 2014; 14(157):1–9.

51. Oğuz F, Unüvar E, Süoğlu Y, Erdamar B, Dündar G, Katircioğlu S, et al. Etiology of acute otitis mediain childhood and evaluation of two different protocols of antibiotic therapy: 10 days cefaclor vs. 3 daysazitromycin. Int J Pediatr Otorhinolaryngol. 2003; 67(1):43–51. PMID: 12560149

52. Guven M, Bulut Y, Sezer T, Aladag I, Eyibilen A, Etikan I. Bacterial etiology of acute otitis media andclinical efficacy of amoxicillin—clavulanate versus azithromycin. International Journal of PediatricOtorhinolaryngology. 2006; 70(70):915–23.

53. Bulut Y, Guven M, Otlu B, Yenisehirli G, Aladag I, Eyibilen A, et al. Acute otitis media and respiratoryviruses. Eur J Pediatr. 2007; 166:223–8. PMID: 16967296

54. Huebner RE, Wasas AD, Hockman M, Klugman KP. Bacterial aetiology of non-resolving otitis mediain South African children. The journal of Laryngology & Otology. 2003; 117:169–72.

55. Couloigner V, Levy C, Francois M, Bidet P, Hausdorff WP, Pascal T, et al. Pathogens implicated inacute otitis media failures after 7-valent pneumococcal conjugate vaccine implementation in France:distribution, serotypes, and resistance levels. Pediatr Infect Dis J. 2012; 31:154–8. doi: 10.1097/INF.0b013e3182357c8d PMID: 21983212

56. Wiertsema SP, Kirkham LS, Corscadden KJ, Mowe EN, Bowman JM, Jacoby P, et al. Predominanceof nontypeable Haemophilus influenzae in children with otitis media following introduction of a 3 + 0pneumococcal conjugate vaccine schedule. Vaccine. 2011; 29:5163–70. doi: 10.1016/j.vaccine.2011.05.035 PMID: 21621576

57. Thornton RB, Wiertsema SP, Kirkham LS, Rigby PJ, Vijayasekaran S, Coates HL, et al. Neutrophilextracellular traps and bacterial biofilms in middle ear effusion of children with recurrent acute otitismedia—a potential treatment target. PLoS One. 2013; 8:e53837. doi: 10.1371/journal.pone.0053837PMID: 23393551

58. Mills N, Best E, Murdoch D, Souter M, Neeff M, Anderson T, et al. What is behind the ear drum? Themicrobiology of otitis media and the nasopharyngeal flora in children in the era of pneumococcal vac-cination. J Paediatr Child Health. 2015; 51(3):300–6. doi: 10.1111/jpc.12710 PMID: 25175818

59. Pereira MBR, Pereira MR, Cantarelli V, Costa SS. Prevalence of bacteria in children with otitis mediawith effusion. Jornal de Pediatria. 2004; 80(1):41–8. PMID: 14978548









http://dx.doi.org/10.1080/00016480802032777




http://dx.doi.org/10.1097/INF.0b013e3182357c8d

http://dx.doi.org/10.1097/INF.0b013e3182357c8d





http://dx.doi.org/10.1371/journal.pone.0053837


http://dx.doi.org/10.1111/jpc.12710



60. Post JC, Preston RA, Aul JJ, Larkins-Pettigrew M, Rydquist-White J, Anderson KW, et al. Molecularanalysis of bacterial pathogens in otitis media with effusion. JAMA. 1995; 273(20):1598–604. PMID:7745773

61. Sipilä P, Jokipii AM, Jokipii L, Karma P. Bacteria in the middle ear and ear canal of patients with secre-tory otitis media and with non-inflamed ears. Acta Otolaryngol. 1981; 92(1–2):123–30. PMID:7315245

62. Sriwardhana KB, Howard AJ, Dunkin KT. Bacteriology of otitis media with effusion. J Laryngol Otol1989; 103(3):253–6. PMID: 2495334

63. Diamond C, Sisson PR, Kearns AM, InghamHR. Bacteriology of chronic otitis media with effusion. JLaryngol Otol. 1989; 103(4):369–71. PMID: 2497219

64. Jero J, Karma P. Bacteriological findings and persistence of middle ear effusion in otitis media witheffusion. Acta Otolaryngol Suppl. 1997; 529:22–6. PMID: 9288259

65. Leskinen K, Hendolin P, Virolainen-Julkunen A, Ylikoski J, Jero J. The clinical role of Alloiococcus oti-tidis in otitis media with effusion. International Journal of Pediatric Otorhinolaryngology. 2002; 66:41–8. PMID: 12363421

66. Hendolin PH, Kärkkäinen U, Himi T, Markkanen A, Ylikoski J. High incidence of Alloiococcus otitis inotitis media with effusion. Pediatr Infect Dis J 1999; 18(10):860–5. PMID: 10530580

67. Martínez IM, Ramos MA, Masgoret PE. Bacterial implication in otitis media with effusion in the child-hood. Acta Otorrinolaringol Esp. 2007; 58(9):408–12. PMID: 17999905

68. Daniel M, Imtiaz-Umer S, Fergie N, Birchall JP, Bayston R. Bacterial involvement in otitis media witheffusion. Int J Pediatr Otorhinolaryngol. 2012; 76(10):1416–22. doi: 10.1016/j.ijporl.2012.06.013PMID: 22819485

69. Gok U, Bulut Y, Keles E, Yalcin S, Doymaz MZ. Bacteriological and PCR analysis of clinical materialaspirated from otitis media with effusions. Int J Pediatr Otorhinolaryngol. 2001; 60(1):49–54. PMID:11434953

70. Jung H, Lee SK, Cha SH, Byun JY, Park MS, Yeo SG. Current bacteriology of chronic otitis mediawith effusion: high rate of nosocomial infection and decreased antibiotic sensitivity. J Infect 2009; 59(5):308–16. doi: 10.1016/j.jinf.2009.08.013 PMID: 19715725

71. Park C-W, Han J-H, Jeong J-H, Cho S-H, Kang M-J, Tea K, et al. Detection rates of bacteria in chronicotitis media with effusion in children. J Korean Med Sci. 2004; 19:735–8. PMID: 15483353

72. Kurono Y, Tomonaga K, Mogi G. Staphylococcus epidermidis and Staphylococcus aureus in otitismedia with effusion. Arch Otolaryngol Head Neck Surg 1988; 114(11):1262–5. PMID: 3262358

73. Shishegar M, Faramarzi A, Kazemi T, Bayat A, Motamedifar M. Polymerase chain reaction, bacterio-logic detection and antibiogram of bacteria isolated from otitis media with effusion in children, shiraz,iran. Iran J Med Sci. 2011; 36(4):273–80. PMID: 23115412

74. Khoramrooz SS, Mirsalehian A, Emaneini M, Jabalameli F, Aligholi M, Saedi B, et al. Frequency ofAlloicoccus otitidis, Streptococcus pneumoniae, Moraxella catarrhalis and Haemophilus influenzae inchildren with otitis media with effusion (OME) in Iranian patients. Auris Nasus Larynx. 2012; 39:369–73. doi: 10.1016/j.anl.2011.07.002 PMID: 21868180

75. Matar GM, Sidani N, Fayad M, Hadi U. Two-step PCR-based assay for identification of bacterial etiol-ogy of otitis media with effusion in infected Lebanese children. J Clin Microbiol. 1998; 36(5):1185–8.PMID: 9574673

76. Nasser SC, Moukarzel N, Nehme A, Haidar H, Kabbara B, Haddad A. Otitis media with effusion in leb-anese children: prevalence and pathogen susceptibility. The journal of Laryngology & Otology. 2011;125:928–33.

77. el-Shamy HA. Bacteriology of chronic secretory otitis media in children. J Egypt Public Health Assoc.1993; 68:495–505. PMID: 7775877

78. Aly BH, Hamad MS, Mohey M, Amen S. Polymerase Chain Reaction (PCR) Versus Bacterial Culturein Detection of Organisms in Otitis Media with Effusion (OME) in Children. Indian J Otolaryngol HeadNeck Surg. 2012; 64(1):51–5. doi: 10.1007/s12070-011-0161-6 PMID: 23449820

79. Stuart J, Butt H, Walker P. The microbiology of glue ear in Australian Aboriginal children. J PaediatrChild Health. 2003; 39(9):665–7. PMID: 14629496

80. Watson P, Voss L, Barber C, Aickin R, Bremner D, Lennon D. The microbiology of chronic otitis mediawith effusion in a group of Auckland children. N Z Med J. 1996; 109(1022):182–4. PMID: 8657383

81. Block SL, Hedrick J, Harrison CJ, Tyler R, Smith A, Findlay R, et al. Pneumococcal serotypes fromacute otitis media in rural Kentucky. Pediatr Infect Dis J. 2002; 21:859–65. PMID: 12352810

82. Bluestone CD, Stephenson JS, Martin LM. Ten-year review of otitis media pathogens. Pediatr InfectDis J. 1992; 11(8):s7–s11. PMID: 1513611














http://dx.doi.org/10.1016/j.jinf.2009.08.013





http://dx.doi.org/10.1016/j.anl.2011.07.002




http://dx.doi.org/10.1007/s12070-011-0161-6






83. Dagan R, Johnson CE, McLinn S, Abughali N, Feris J, Leibovitz E, et al. Bacteriologic and clinical effi-cacy of amoxicillin/clavulanate vs. azithromycin in acute otitis media. Pediatr Infect Dis J. 2000; 19(2):95–104. PMID: 10693993

84. Chiu NC, Lin HY, Hsu CH, Huang FY, Lee KS, Chi H. Epidemiological and microbiological character-istics of culture-proven acute otitis media in Taiwanese children. J Formos Med Assoc. 2012; 111(10):536–41. doi: 10.1016/j.jfma.2011.07.015 PMID: 23089688

85. Leibovitz E, Piglansky L, Raiz S, Greenberg D, Hamed KA, Ledeine JM, et al. Bacteriologic and clini-cal efficacy of oral gatifloxacin for the treatment of recurrent/nonresponsive acute otitis media: anopen label, noncomparative, double tympanocentesis study. Pediatr Infect Dis J 2003; 22(11):943–9.PMID: 14614364

86. Ben-Shimol S, Greenberg D, Givon-Lavi N, Elias N, Glikman D, Rubinstein U, et al. Rapid reduction ininvasive pneumococcal disease after introduction of PCV7 into the National Immunization Plan inIsrael. Vaccine. 2012; 30(46):6600–7. doi: 10.1016/j.vaccine.2012.08.012 PMID: 22939907

87. Jones SA, GroomeM, Koen A, Van NN, Sewraj P, Kuwanda L, et al. Immunogenicity of seven-valentpneumococcal conjugate vaccine administered at 6, 14 and 40 weeks of age in South african infants.PLoS One. 2013; 8(8):e72794. doi: 10.1371/journal.pone.0072794 PMID: 24015277

88. Abdelnour A, Soley C, Guevara S, Porat N, Dagan R, Arguedas A. Streptococcus pneumoniae Sero-type 3 among Costa Rican Children with Otitis Media: clinical, epidemiological characteristics andantimicrobial resistance patterns. BMC Pediatrics. 2009; 9(52):1–6.

89. Lee LH, Gu X, NahmMH. Towards New Broader Spectrum Pneumococcal Vaccines: The Future ofPneumococcal Disease Prevention. Vaccines. 2014; 2:112–28. doi: 10.3390/vaccines2010112PMID: 26344470

90. Gomez-Barreto D, Monteros LEEdl, Lopez-Enriquez C, Suarez RR, Torre Cdl. Streptococcus pneu-moniae serotypes isolated frommiddle ear of Mexican children diagnosed with acute otitis media.salud pública de méxico. 2011; 53(3):207–11. PMID: 21829885

91. Guevara S, Abdelnour A, Soley C, Porat N, Dagan R, Arguedas A. Streptococcus pneumoniae sero-types isolated from the middle ear fluid of Costa Rican children following introduction of the heptava-lent pneumococcal conjugate vaccine into a limited population. Vaccine. 2012; 30(26):3857–61. doi:10.1016/j.vaccine.2012.04.010 PMID: 22521846

92. Arguedas A, Dagan R, Guevara S, Porat N, Soley C, Perez A, et al. Middle ear fluid Streptococcuspneumoniae serotype distribution in Costa Rican children with otitis media. The Pediatric InfectiousDisease Journal. 2005; 24(7):631–4. PMID: 15999006

93. Centers for Disease Control and Prevention. Preventing Pneumococcal Disease Among Infants andYoung Children. 2000.

94. McEllistremMC, Adams JM, Patel K, Mendelsohn AB, Kaplan SL, Bradley JS, et al. Acute otitis mediadue to penicillin-nonsusceptible Streptococcus pneumoniae before and after the introduction of thepneumococcal conjugate vaccine. Clin Infect Dis. 2005; 40(12):1738–44. PMID: 15909260

95. Gomes HDC, Muscat M, Monnet DL, Giesecke J, Lopalco PL. Use of seven-valent pneumococcalconjugate vaccine (PCV7) in Europe, 2001–2007. EUROSURVEILLANCE. 2009; 14(12):1–6.

96. Stamboulidis K, Chatzaki D, Poulakou G, Ioannidou S, Lebessi E, Katsarolis I, et al. The impact of theheptavalent pneumococcal conjugate vaccine on the epidemiology of acute otitis media complicatedby otorrhea. Pediatr Infect Dis J. 2011; 30:551–5. doi: 10.1097/INF.0b013e31821038d9 PMID:21297521

97. Gene A, Garcia-Garcia JJ, Domingo A, Palacin PWyE. Etiology of acute otitis media in a children'shospital and antibiotic sensitivity of the bacteria involved. Enferm Infecc Microbiol Clin. 2004; 22(7):377–80. PMID: 15355766

98. Fenoll A, Aguilar L, Vicioso M-D, Robledo O, Granizo J-J. Increase in serotype 19A prevalence andamoxicillin non-susceptibility among paediatric Streptococcus pneumoniae isolates frommiddle earfluid in a passive laboratory-based surveillance in Spain, 1997–2009. BMC Infectious Diseases.2011; 11(239):1–8.

99. Alonso M, Marimon JM, ErcibengoaM, Pérez-Yarza EG, Pérez-Trallero E. Dynamics of Streptococ-cus pneumoniae serotypes causing acute otitis media isolated from children with spontaneous mid-dle-ear drainage over a 12-year period (1999–2010) in a region of northern Spain. PLoS One. 2013; 8(1):e54333. doi: 10.1371/journal.pone.0054333 PMID: 23349853

100. Lu C-Y, SantoshamM, Members A. Survey of national immunization programmes and vaccine cover-age rates in Asia Pacific countries. Vaccine. 2012; 30:2250–5. doi: 10.1016/j.vaccine.2011.10.070PMID: 22075085

101. Tyo KR, Rosen MM, ZengW, Yap M, Pwee KH, Ang LW, et al. Cost-effectiveness of conjugate pneu-mococcal vaccination in Singapore: comparing estimates for 7-valent, 10-valent, and 13-valent vac-cines. Vaccine. 2011; 29:6686–94. doi: 10.1016/j.vaccine.2011.06.091 PMID: 21745516




http://dx.doi.org/10.1016/j.jfma.2011.07.015







http://dx.doi.org/10.3390/vaccines2010112







http://dx.doi.org/10.1097/INF.0b013e31821038d9









102. Porat N, Amit U, Givon-Lavi N, Leibovitz E, Dagan R. Increasing Importance of Multidrug-ResistantSerotype 6A Streptococcus pneumoniae Clones in Acute Otitis Media in Southern Israel. The Pediat-ric Infectious Disease Journal. 2010; 29(2):126–30. doi: 10.1097/INF.0b013e3181b78e6e PMID:19927039

103. Somech I, Dagan R, Givon-Lavi N, Porat N, Raiz S, Leiberman A, et al. Distribution, dynamics andantibiotic resistance patterns of Streptococcus pneumoniae serotypes causing acute otitis media inchildren in southern Israel during the 10 year-period before the introduction of the 7-valent pneumo-coccal conjugate vaccine. Vaccine. 2011; 29(25):4202–9. doi: 10.1016/j.vaccine.2011.03.103 PMID:21497634

104. Kim SY, Lee G, Goldie SJ. Economic evaluation of pneumococcal conjugate vaccination in The Gam-bia. BMC Infect Dis. 2010; 10(260):1–18.

105. Ako-Nai AK, Oluga FA, Onipede AO, Adejuyigbe EA, Amusa YB. The characterization of bacterial iso-lates from acute otitis media in Ile-Ife, southwestern Nigeria. Journal of Tropical Pediatrics. 2002;48:15–23. PMID: 11866331

106. Akinjogunla OJ, Eghafona NO, Enabulele IO. Aetiologic Agents of Acute Otitis Media (AOM): Preva-lence, Antibiotic Susceptibility, β-Lactamase (βL) and Extended Spectrum β-Lactamase (ESBL) Pro-duction. Journal of Microbiology, Biotechnology and Food Sciences. 2011; 1(3):333–53.

107. Giele C, Moore H, Bayley K, Harrison C, Murphy D, Rooney K, et al. Has the seven-valent pneumo-coccal conjugate vaccine had an impact on invasive pneumococcal disease in Western Australia?Vaccine. 2007; 25(13):2379–84. PMID: 17064825

108. Watson M, Brett M, BrownM, Stewart MG, Warren S, Network ftNSWP. Pneumococci responsible forinvasive disease and discharging ears in children in Sydney, Australia. journal of medical microbiol-ogy. 2007; 56:819–23. PMID: 17510269

109. Dawson VM, Coelen RJ, Murphy S, Graham D, Dyer H, Sunderman J. Microbiology of chronic otitismedia with effusion among Australian Aboriginal children: role of Chlamydia trachomatis. Aust J ExpBiol Med Sci. 1985; 63:99–107. PMID: 4040361

110. Gibney KB, Morris PS, Carapetis JR, Skull SA, Smith-Vaughan HS, Stubbs E, et al. The clinicalcourse of acute otitis media in high-risk Australian Aboriginal children: a longitudinal study. BMCPediatr. 2005; 5(16):1–8.

111. Leach AJ, Morris PS. The burden and outcome of respiratory tract infection in Australian and Aborigi-nal children. Pediatr Infect Dis J. 2007; 26(10):S4–S7. PMID: 18049380

112. Walls T, Best E, Murdoch D, Mills N. Vaccination to prevent otitis media in New Zealand. N Z Med J.2011; 124(1340):6–9. PMID: 21952379

113. Hamamoto Y, Gotoh Y, Nakajo Y, Shimoya S, Kayama C, Hasegawa S, et al. Impact of antibiotics onpathogens associated with otitis media with effusion. J Laryngol Otol. 2005; 119(11):862–5. PMID:16354337

114. Kalcioglu MT, Oncel S, Durmaz R, Otlu B, Miman MC, Ozturan O. Bacterial etiology of otitis mediawith effusion; focusing on the high positivity of Alloiococcus otitidis. NewMicrobiol. 2002; 25(1):31–5.PMID: 11837388

115. Neumark T, EkblomM, Brudin L, Groth A, Eliasson I, Molstad S, et al. Spontaneously draining acuteotitis media in children: An observational study of clinical findings, microbiology and clinical course.Scandinavian Journal of Infectious Diseases. 2011; 43:891–8. doi: 10.3109/00365548.2011.591820PMID: 21736512

116. Beswick AJ, Lawley B, Fraise AP, Pahor AL, Brown NL. Detection of Alloiococcus otitis in mixed bac-terial populations frommiddle-ear effusions of patients with otitis media. Lancet. 1999; 354(9176):386–9. PMID: 10437868

117. Holder RC, Kirse DJ, Evans AK, Peters TR, Poehling KA, SwordsWE, et al. One third of middle eareffusions from children undergoing tympanostomy tube placement had multiple bacterial pathogens.BMC Pediatrics. 2012; 12(87):1–7.

118. Güvenç MG, Midilli K, Inci E, Kuşkucu M, Tahamiler R, Ozergil E, et al. Lack of Chlamydophila pneu-moniae and predominance of Alloiococcus otitidis in middle ear fluids of children with otitis media witheffusion. Auris Nasus Larynx. 2010; 37(3):269–73. doi: 10.1016/j.anl.2009.09.002 PMID: 19879704

119. Ashhurst-Smith C, Hall ST, Walker P, Stuart J, Hansbro PM, Blackwell CC. Isolation of Alloiococcusotitidis from Indigenous and non-Indigenous Australian children with chronic otitis media with effusion.FEMS immunology and medical microbiology. 2007; 51(1):163–70. PMID: 17666076

120. Morris PS, Leach AJ. acute and chronic otitis media. Pediatr Clin N Am. 2009; 56:1383–99.

121. Ashhurst-Smith C, Hall S, Burns C, Stuart J, Blackwell C. Induction of inflammatory responses fromTHP-1 cells by cell-free filtrates from clinical isolates of Alloiococcus otitidis. Innate Immun. 2014; 20(3).



http://dx.doi.org/10.1097/INF.0b013e3181b78e6e












http://dx.doi.org/10.3109/00365548.2011.591820



http://dx.doi.org/10.1016/j.anl.2009.09.002



122. Ashhurst-Smith C, Hall ST, Burns CJ, Stuart J, Blackwell CC. In vitro inflammatory responses elicitedby isolates of Alloiococcus otitidis obtained from children with otitis media with effusion. InnateImmun. 2014; 3:Abstract.

123. Palmu AA, Jokinen JT, Kaijalainen T, Leinonen M, Karma P, Kilpi TM. Association of clinical signsand symptoms with pneumococcal acute otitis media by serotype—implications for vaccine effect.Clin Infect Dis. 2005; 40(1):52–7. PMID: 15614692

124. RodriguezWJ, Schwartz RH. Streptococcus pneumoniae causes otitis media with higher fever andmore redness of tympanic membranes than Haemophilus influenzae or Moraxella catarrhalis. PediatrInfect Dis J. 1999; 18(10):942–4. PMID: 10530598

125. Courter JD, Baker WL, Nowak KS, Smogowicz LA, Desjardins LL, Coleman CI, et al. Increased clini-cal failures when treating acute otitis media with macrolides: a meta-analysis. Ann Pharmacother2010; 44:471–8. doi: 10.1345/aph.1M344 PMID: 20150506

126. Pichichero ME, Casey JR. Emergence of a Multiresistant Serotype 19A Pneumococcal Strain NotIncluded in the 7-Valent Conjugate Vaccine as an Otopathogen in Children. JAMA. 2007; 298(15):1772–8. PMID: 17940232

127. Nuorti JP, Whitney CG. Prevention of Pneumococcal Disease Among Infants and Children—Use of13-Valent Pneumococcal Conjugate Vaccine and 23-Valent Pneumococcal Polysaccharide Vaccine:Recommendations of the Advisory Committee on Immunization Practices (ACIP). Morbidity and Mor-tality Weekly Report (CDC). 2010; 59:1–18.

128. Jefferies JMC, Macdonald E, Faust SN, Clarke SC. 13-valent pneumococcal conjugate vaccine(PCV13). Human Vaccines. 2011; 7(10):1012–8. doi: 10.4161/hv.7.10.16794 PMID: 21941097

129. Newall AT, Creighton P, Philp DJ, Wood JG, MacIntyre CR. The potential cost-effectiveness of infantpneumococcal vaccines in Australia. Vaccine. 2011; 29:8077–85. doi: 10.1016/j.vaccine.2011.08.050PMID: 21864617

130. Smith-Vaughan HC, Binks MJ, Marsh RL, Kaestli M, Ward L, Hare KM, et al. Dominance of Haemo-philus influenzae in ear discharge from Indigenous Australian children with acute otitis media with tym-panic membrane perforation. BMC Ear Nose Throat Disord. 2013; 13(1):1–9.

131. Palmu AA, Herva E, Savolainen H, Karma P, Mäkelä PH, Kilpi TM. Association of clinical signs andsymptoms with bacterial findings in acute otitis media. Clin Infect Dis. 2004; 38(2):234–42. PMID:14699456

132. Barkai G, Leibovitz E, Givon-Lavi N, Dagan R. Potential contribution by nontypable Haemophilusinfluenzae in protracted and recurrent acute otitis media. Pediatr Infect Dis J. 2009; 28:466–71. PMID:19504729

133. Livermore DM. b-Lactamases in Laboratory and Clinical Resistance. CLINICAL MICROBIOLOGYREVIEWS. 1995; 8(4):557–84. PMID: 8665470

134. Wilke MS, Lovering AL, Strynadka NCJ. b-Lactam antibiotic resistance: a current structural perspec-tive. Current Opinion in Microbiology. 2005; 8:525–33. PMID: 16129657

135. Yano H, Yamazaki Y, Qin L, Okitsu N, Yahara K, Irimada M, et al. Improvement rate of acute otitismedia caused by Haemophilus influenzae at 1 week is significantly associated with time to recovery.J Clin Microbiol. 2013; 51(11):3542–6. doi: 10.1128/JCM.01108-13 PMID: 23966504

136. Tristram S, Jacobs M, Appelbaum P. Antimicrobial resistance in Haemophilus influenzae. Clin Micro-biol Rev. 2007; 20(2):368–89. PMID: 17428889

137. Tomic V, Dowzicky M. Regional and global antimicrobial susceptibility among isolates of Streptococ-cus pneumoniae and Haemophilus influenzae collected as part of the Tigecycline Evaluation and Sur-veillance Trial (T.E.S.T.) from 2009 to 2012 and comparison with previous years of T.E.S.T. (2004–2008). Ann Clin Microbiol Antimicrob 2014; 13(52):1–8.

138. Agrawal A, Murphy TF. Haemophilus influenzae Infections in the H. influenzae Type b Conjugate Vac-cine Era. JOURNALOF CLINICALMICROBIOLOGY. 2011; 49(11):3728–32. doi: 10.1128/JCM.05476-11 PMID: 21900515

139. Murphy TF. Vaccine development for non-typeable Haemophilus influenzae and Moraxella catarrha-lis: progress and challenges. Expert Rev Vaccines. 2005; 4(6):843–53. PMID: 16372880

140. Ruiz-Palacios GM, Guerrero ML, Hernández-Delgado L, Lavalle-Villalobos A, Casas-Muñoz A, Cer-vantes-Apolinar Y, et al. Immunogenicity, reactogenicity and safety of the 10-valent pneumococcalnontypeable Haemophilus influenzae protein D conjugate vaccine (PHiD-CV) in Mexican infants.Hum Vaccin. 2011; 7(11):1137–45. doi: 10.4161/hv.7.11.17984 PMID: 22048109

141. Lagos RE, Muñoz AE, Levine MM, Lepetic A, François N, Yarzabal JP, et al. Safety and immunoge-nicity of the 10-valent pneumococcal nontypeable Haemophilus influenzae protein D conjugate vac-cine (PHiD-CV) in Chilean children. Hum Vaccin. 2011; 7(5):511–22. PMID: 21441782





http://dx.doi.org/10.1345/aph.1M344



http://dx.doi.org/10.4161/hv.7.10.16794








http://dx.doi.org/10.1128/JCM.01108-13







http://dx.doi.org/10.4161/hv.7.11.17984



142. Dicko A, Santara G, Mahamar A, Sidibe Y, Barry A, Dicko Y, et al. Safety, reactogenicity and immuno-genicity of a booster dose of the 10-valent pneumococcal non-typeable Haemophilus influenzae pro-tein D conjugate vaccine (PHiD-CV) in Malian children. Hum Vaccin Immunother. 2013; 9(2):382–8.PMID: 23291945

143. van den Bergh MR, Spijkerman J, François N, Swinnen K, Borys D, Schuerman L, et al. Immunoge-nicity, safety, and reactogenicity of the 10-valent pneumococcal nontypeable Haemophilus influenzaeprotein D conjugate vaccine and DTPa-IPV-Hib when coadministered as a 3-dose primary vaccinationschedule in The Netherlands: a randomized controlled trial. Pediatr Infect Dis J 2011; 30(9):e179–e8.

144. Kim CH, Kim JS, Cha SH, Kim KN, Kim JD, Lee KY, et al. Response to primary and booster vaccina-tion with 10-valent pneumococcal nontypeable Haemophilus influenzae protein D conjugate vaccinein Korean infants. Pediatr Infect Dis J. 2011; 30(12):e235–e43. doi: 10.1097/INF.0b013e31822a8541PMID: 21817957

145. Lim FS, Koh MT, Tan KK, Chan PC, Chong CY, Shung YYW, et al. A randomised trial to evaluate theimmunogenicity, reactogenicity, and safety of the 10-valent pneumococcal non-typeable Haemophi-lus influenzae protein D conjugate vaccine (PHiD-CV) co-administered with routine childhood vac-cines in Singapore and Malaysia. BMC Infect Dis. 2014; 14(530):1–13.

146. Smith-Vaughan HC, Chang AB, Sarovich DS, Marsh RL, Grimwood K, Leach AJ, et al. Absence of animportant vaccine and diagnostic target in carriage- and disease-related nontypeable Haemophilusinfluenzae. Clin Vaccine Immunol. 2014; 21(2):250–2. doi: 10.1128/CVI.00632-13 PMID: 24285816

147. Verduin CM, Hol C, Fleer A, Dijk Hv, van Belkum A. Moraxella catarrhalis: from Emerging to Estab-lished Pathogen. CLINICAL MICROBIOLOGY REVIEWS. 2002; 15(1):125–44. PMID: 11781271

148. Coffey JDJ, Martin AD, Booth HN. Neisseria catarrhalis in exudate otitis media. Arch Otolaryngol.1967; 86(4):403–6. PMID: 6041112

149. Broides A, Dagan R, Greenberg D, Givon-Lavi N, Leibovitz E. Acute otitis media caused by Moraxellacatarrhalis: epidemiologic and clinical characteristics. Clin Infect Dis 2009; 49:1641–7. doi: 10.1086/647933 PMID: 19886799

150. Tan TT, Riesbeck K. Current progress of adhesins as vaccine candidates for Moraxella catarrhalis.Expert Rev Vaccines. 2007; 6(6):949–56. doi: 10.1586/14760584.6.6.949 PMID: 18377357

151. Mawas F, Ho MM, Corbel MJ. Current progress with Moraxella catarrhalis antigens as vaccine candi-dates. Expert Rev Vaccines. 2009; 8(1):77–90. doi: 10.1586/14760584.8.1.77 PMID: 19093775

152. Smidt M, Bättig P, Verhaegh SJ, Niebisch A, Hanner M, Selak S, et al. Comprehensive antigenscreening identifies Moraxella catarrhalis proteins that induce protection in a mouse pulmonary clear-ance model. PLoS One. 2013; 8(5):1–15.

153. Segal N, Givon-Lavi N, Leibovitz E, Yagupsky P, Leiberman A, Dagan R. Acute Otitis Media Causedby Streptococcus pyogenes in Children. Clinical Infectious Diseases. 2005; 41:35–41. PMID:15937760

154. Roland PS. controversial etiopathologic agents in otitis media. Ear, Nose & Throat Journal. 2007; 86(11):2–4.

155. Harimaya A, Fujii N, Himi T. Preliminary study of proinflammatory cytokines and chemokines in themiddle ear of acute otitis media due to Alloiococcus otitidis. Int J Pediatr Otorhinolaryngol. 2009; 73(5):677–80. doi: 10.1016/j.ijporl.2008.12.033 PMID: 19185927

156. Marsh RL, Binks MJ, Beissbarth J, Christensen P, Morris PS, Leach AJ, et al. Quantitative PCR of eardischarge from indigenous Australian children with acute otitis media with perforation supports a rolefor Alloiococcus otitidis as a secondary pathogen. BMC Ear, Nose and Throat Disorders. 2012; 12(11).

157. Cripps AW, Otczyk DC. Prospects for a vaccine against otitis media. Expert Rev Vaccines. 2006; 5(4):517–34. PMID: 16989632

158. Niittynen L, Pitkäranta A, Korpela R. Probiotics and otitis media in children. Int J Pediatr Otorhinolaryn-gol. 2012; 76(4):465–70. doi: 10.1016/j.ijporl.2012.01.011 PMID: 22305688




http://dx.doi.org/10.1097/INF.0b013e31822a8541


http://dx.doi.org/10.1128/CVI.00632-13




http://dx.doi.org/10.1086/647933

http://dx.doi.org/10.1086/647933


http://dx.doi.org/10.1586/14760584.6.6.949


http://dx.doi.org/10.1586/14760584.8.1.77








293

Appendix B:

Media and solutions

294

Media

Skim milk tryptone glucose glycerol broth (STGGB) for sample

collection

30g/L Tryptone Soy Broth (Oxoid)

5g/L Glucose (Sigma)

20g/L Skim milk powder (Oxoid)

10% v/v Glycerol (Merck)

Distilled water

Brain Heart Infusion (BHI) broth for Spn and Mcat culture

37g/L BHI broth (Oxoid)

Distilled water

Haemophilus Test Medium (HTM) broth for Hi culture

22g/L Mueller-Hinton broth (Oxoid)

4g/L Yeast extract (Oxoid)

Distilled water

1 vial of HTM supplement (Oxoid)

Blood BHI agar for Mcat culture

52g/L BHI agar (Oxoid)

Distilled water

5% v/v defibrinated horse blood (Oxoid)

295

Blood Trypticase soy agar (TSA) for Spn culture

40g/L TSA (Oxoid)

Distilled water


Chocolate agar for Hi culture

72g/L GC agar base (Oxoid)

Distilled water


Heat at 70oC until the colour changing from bright to dark red

(chocolate colour)

Solutions

0.5M Ethylenediaminetetraacetic acid (EDTA)

186.1g/L EDTA disodium salt dehydrate (Sigma)

Distilled water

5M Sodium chloride (NaCl)

292.2g/L NaCl (Sigma)

Distilled water

3x Phosphate-buffered saline (PBS), pH 7.2

22.8g/L NaCl

1.16g/L Sodium phosphate monobasic (NaH2PO4) (Sigma)

3.8g/L Sodium phosphate dibasic (Na2HPO4) (Sigma)

Distilled water

296

1x Tris(hydroxymethyl)aminomethane (Tris) and Hydrogen chloride

(HCl) (Tris – HCl), pH8

121.1g/L Tris (Sigma)

Distilled water

Concentrated HCl was used for pH adjustment

4% Paraformaldehyde (PFA)

40g/L PFA (Sigma)

1x PBS

10% Sodium dodecyl sulfate (SDS), pH 7.2

100g/L SDS (Sigma)

Distilled water

1M Trizma base

121.1g/L Trizma base (Sigma)

Distilled water

1M Sodium dihydrogenphosphate (NaH2PO4.2H2O)

156.2g/L NaH2PO4.2H2O (Sigma)

Distilled water

Tris-PO4 buffer, pH 9.0

Titrate 50ml of Trizma with the NaH2PO4.2H2O drop wise to reach

pH 9.0.

297

Low fade mounting media

(Recipe provided by CMCA, University of Western Australia)

5ml Tris-PO4 buffer

20g Polyvinyl alcohol (PVA) (Sigma)

30ml glycerol (Sigma)

100mg Chlorobutanol (Sigma)

75ml distilled water

Mix 5ml Tris-PO4 buffer and 75ml H2O in Erlenmeyer flask. Add PVA to the

solution and place the flask in a 60oC oven for 2-3h. Slowly add glycerol

and Chlorobutanol. Adjust to pH8.2 with Tris-PO4. Cover the media with

aluminium foil and store at 4oC.

298

Appendix C:

List of genes in Nanostring analysis

299

Official Symbol Official Full Name

ABCB1 ATP-binding cassette, sub-family B (MDR/TAP), member 1

ABL1 c-abl oncogene 1, non-receptor tyrosine kinase

ADA adenosine deaminase

AHR aryl hydrocarbon receptor

AICDA activation-induced cytidine deaminase

AIRE autoimmune regulator

APP amyloid beta (A4) precursor protein

ARG1 arginase, liver

ARG2 arginase, type II

ARHGDIB Rho GDP dissociation inhibitor (GDI) beta

ATG10 ATG10 autophagy related 10 homolog (S. cerevisiae)


ATG16L1 ATG16 autophagy related 16-like 1 (S. cerevisiae)



ATM ataxia telangiectasia mutated

B2M beta-2-microglobulin

B3GAT1 beta-1,3-glucuronyltransferase 1 (glucuronosyltransferase P)

BATF basic leucine zipper transcription factor, ATF-like

BATF3 basic leucine zipper transcription factor, ATF-like 3

BAX BCL2-associated X protein

BCAP31 B-cell receptor-associated protein 31

BCL10 B-cell CLL/lymphoma 10


BCL2L11 BCL2-like 11 (apoptosis facilitator)



BID BH3 interacting domain death agonist

BLNK B-cell linker

BST1 bone marrow stromal cell antigen 1

BST2 bone marrow stromal cell antigen 2

BTK Bruton agammaglobulinemia tyrosine kinase

BTLA B and T lymphocyte associated

C14orf166 chromosome 14 open reading frame 166

C1QA complement component 1, q subcomponent, A chain

C1QB complement component 1, q subcomponent, B chain

C1QBP complement component 1, q subcomponent binding protein

C1R complement component 1, r subcomponent

C1S complement component 1, s subcomponent

C2 complement component 2


C4A/B complement component 4A (Rodgers blood group)/complement component 4B (Chido blood group)

C4BPA complement component 4 binding protein, alpha



300


C8A complement component 8, alpha polypeptide

C8B complement component 8, beta polypeptide

C8G complement component 8, gamma polypeptide


CAMP cathelicidin antimicrobial peptide

CARD9 caspase recruitment domain family, member 9

CASP1 caspase 1, apoptosis-related cysteine peptidase (interleukin 1, beta, convertase)

CASP10 caspase 10, apoptosis-related cysteine peptidase




CCBP2 chemokine binding protein 2

CCL11 chemokine (C-C motif) ligand 11




CCL18 chemokine (C-C motif) ligand 18 (pulmonary and activation-regulated)













CCND3 cyclin D3

CCR1 chemokine (C-C motif) receptor 1







CCRL1 chemokine (C-C motif) receptor-like 1

CCRL2 chemokine (C-C motif) receptor-like 2

CD14 CD14 molecule

CD160 CD160 molecule


CD164 CD164 molecule, sialomucin

CD19 CD19 molecule

301

CD1A CD1a molecule

CD1D CD1d molecule

CD2 CD2 molecule


CD22 CD22 molecule

CD24 CD24 molecule

CD244 CD244 molecule, natural killer cell receptor 2B4


CD27 CD27 molecule



CD28 CD28 molecule

CD34 CD34 molecule

CD36 CD36 molecule (thrombospondin receptor)

CD3D CD3d molecule, delta (CD3-TCR complex)

CD3E CD3e molecule, epsilon (CD3-TCR complex)

CD3EAP CD3e molecule, epsilon associated protein

CD4 CD4 molecule

CD40 CD40 molecule, TNF receptor superfamily member 5

CD40LG CD40 ligand

CD44 CD44 molecule (Indian blood group)

CD46 CD46 molecule, complement regulatory protein

CD48 CD48 molecule

CD5 CD5 molecule

CD53 CD53 molecule

CD55 CD55 molecule, decay accelerating factor for complement (Cromer blood group)

CD58 CD58 molecule

CD59 CD59 molecule, complement regulatory protein

CD6 CD6 molecule

CD7 CD7 molecule

CD70 CD70 molecule

CD74 CD74 molecule, major histocompatibility complex, class II invariant chain

CD79A CD79a molecule, immunoglobulin-associated alpha

CD79B CD79b molecule, immunoglobulin-associated beta

CD80 CD80 molecule

CD81 CD81 molecule

CD82 CD82 molecule

CD83 CD83 molecule

CD86 CD86 molecule

CD8A CD8a molecule

CD8B CD8b molecule

CD9 CD9 molecule

CD96 CD96 molecule

CD97 CD97 molecule

CD99 CD99 molecule

302

CDH5 cadherin 5, type 2 (vascular endothelium)

CDKN1A cyclin-dependent kinase inhibitor 1A (p21, Cip1)

CEACAM1 carcinoembryonic antigen-related cell adhesion molecule 1 (biliary glycoprotein)

CEACAM6 carcinoembryonic antigen-related cell adhesion molecule 6 (non-specific cross reacting antigen)

CEACAM8 carcinoembryonic antigen-related cell adhesion molecule 8

CEBPB CCAAT/enhancer binding protein (C/EBP), beta

CFB complement factor B

CFD complement factor D (adipsin)

CFH complement factor H

CFI complement factor I

CFP complement factor properdin

CHUK conserved helix-loop-helix ubiquitous kinase

CIITA class II, major histocompatibility complex, transactivator

CISH cytokine inducible SH2-containing protein


CLEC4E C-type lectin domain family 4, member E




CLU clusterin

CMKLR1 chemokine-like receptor 1

CR1 complement component (3b/4b) receptor 1 (Knops blood group)

CR2 complement component (3d/Epstein Barr virus) receptor 2

CRADD CASP2 and RIPK1 domain containing adaptor with death domain

CSF1 colony stimulating factor 1 (macrophage)

CSF1R colony stimulating factor 1 receptor

CSF2 colony stimulating factor 2 (granulocyte-macrophage)

CSF2RB colony stimulating factor 2 receptor, beta, low-affinity (granulocyte-macrophage)

CSF3R colony stimulating factor 3 receptor (granulocyte)

CTLA4_all (common probe) cytotoxic T-lymphocyte-associated protein 4

CTLA4-TM (membrane-bound form) cytotoxic T-lymphocyte-associated protein 4

sCTLA4 (soluble form) cytotoxic T-lymphocyte-associated protein 4

CTNNB1 catenin (cadherin-associated protein), beta 1, 88kDa

CTSC cathepsin C

CTSG cathepsin G

CTSS cathepsin S

CUL9 cullin 9

CX3CL1 chemokine (C-X3-C motif) ligand 1

CX3CR1 chemokine (C-X3-C motif) receptor 1

CXCL1 chemokine (C-X-C motif) ligand 1 (melanoma growth stimulating activity, alpha)

CXCL10 chemokine (C-X-C motif) ligand 10


303





CXCR1 chemokine (C-X-C motif) receptor 1





CYBB cytochrome b-245, beta polypeptide

DEFB1 defensin, beta 1

DEFB103A defensin, beta 103A

DEFB103B defensin, beta 103B

DEFB4A defensin, beta 4A

DPP4 dipeptidyl-peptidase 4

DUSP4 dual specificity phosphatase 4

EBI3 Epstein-Barr virus induced 3

EDNRB endothelin receptor type B

EGR1 early growth response 1

EGR2 early growth response 2

ENTPD1 ectonucleoside triphosphate diphosphohydrolase 1

EOMES eomesodermin

ETS1 v-ets erythroblastosis virus E26 oncogene homolog 1 (avian)

FADD Fas (TNFRSF6)-associated via death domain

FAS Fas (TNF receptor superfamily, member 6)

FCAR Fc fragment of IgA, receptor for

FCER1A Fc fragment of IgE, high affinity I, receptor for; alpha polypeptide

FCER1G Fc fragment of IgE, high affinity I, receptor for; gamma polypeptide

FCGR1A/B Fc fragment of IgG, high affinity Ia, receptor (CD64)/Fc fragment of IgG, high affinity Ib, receptor (CD64)

FCGR2A Fc fragment of IgG, low affinity IIa, receptor (CD32)

FCGR2A/C Fc fragment of IgG, low affinity IIa, receptor (CD32)/Fc fragment of IgG, low affinity IIc, receptor for (CD32)

FCGR2B Fc fragment of IgG, low affinity IIb, receptor (CD32)

FCGR3A/B Fc fragment of IgG, low affinity IIIa, receptor (CD16a)/Fc fragment of IgG, low affinity IIIb, receptor (CD16a)

FCGRT Fc fragment of IgG, receptor, transporter, alpha

FKBP5 FK506 binding protein 5

FN1 fibronectin 1

FOXP3 forkhead box P3

FYN FYN oncogene related to SRC, FGR, YES

GATA3 GATA binding protein 3

GBP1 guanylate binding protein 1, interferon-inducible

GBP5 guanylate binding protein 5

GFI1 growth factor independent 1 transcription repressor

GNLY granulysin

GP1BB glycoprotein Ib (platelet), beta polypeptide

GPI glucose-6-phosphate isomerase

304

GPR183 G protein-coupled receptor 183

GZMA granzyme A (granzyme 1, cytotoxic T-lymphocyte-associated serine esterase 3)

GZMB granzyme B (granzyme 2, cytotoxic T-lymphocyte-associated serine esterase 1)

GZMK granzyme K (granzyme 3; tryptase II)

HAMP hepcidin antimicrobial peptide

HAVCR2 hepatitis A virus cellular receptor 2

HFE hemochromatosis

HLA-A major histocompatibility complex, class I, A

HLA-B major histocompatibility complex, class I, B

HLA-C major histocompatibility complex, class I, C

HLA-DMA major histocompatibility complex, class II, DM alpha

HLA-DMB major histocompatibility complex, class II, DM beta

HLA-DOB major histocompatibility complex, class II, DO beta

HLA-DPA1 major histocompatibility complex, class II, DP alpha 1

HLA-DPB1 major histocompatibility complex, class II, DP beta 1

HLA-DQA1 major histocompatibility complex, class II, DQ alpha 1

HLA-DQB1 major histocompatibility complex, class II, DQ beta 1

HLA-DRA major histocompatibility complex, class II, DR alpha

HLA-DRB1 major histocompatibility complex, class II, DR beta 1

HLA-DRB3 major histocompatibility complex, class II, DR beta 3

HRAS v-Ha-ras Harvey rat sarcoma viral oncogene homolog

ICAM1 intercellular adhesion molecule 1



ICAM4 intercellular adhesion molecule 4 (Landsteiner-Wiener blood group)

ICAM5 intercellular adhesion molecule 5, telencephalin

ICOS inducible T-cell co-stimulator

ICOSLG inducible T-cell co-stimulator ligand

IDO1 indoleamine 2,3-dioxygenase 1

IFI16 interferon, gamma-inducible protein 16

IFI35 interferon-induced protein 35

IFIH1 interferon induced with helicase C domain 1

IFIT2 interferon-induced protein with tetratricopeptide repeats 2

IFITM1 interferon induced transmembrane protein 1 (9-27)

IFNA1/13 interferon, alpha 1/interferon, alpha 13

IFNA2 interferon, alpha 2

IFNAR1 interferon (alpha, beta and omega) receptor 1

IFNAR2 interferon (alpha, beta and omega) receptor 2

IFNB1 interferon, beta 1, fibroblast

IFNG interferon, gamma

IFNGR1 interferon gamma receptor 1

IGF2R insulin-like growth factor 2 receptor

IKBKAP inhibitor of kappa light polypeptide gene enhancer in B-cells, kinase complex-associated protein

IKBKB inhibitor of kappa light polypeptide gene enhancer in B-cells, kinase beta

305

IKBKE inhibitor of kappa light polypeptide gene enhancer in B-cells, kinase epsilon

IKBKG inhibitor of kappa light polypeptide gene enhancer in B-cells, kinase gamma

IKZF1 IKAROS family zinc finger 1 (Ikaros)

IKZF2 IKAROS family zinc finger 2 (Helios)

IKZF3 IKAROS family zinc finger 3 (Aiolos)

IL10 interleukin 10

IL10RA interleukin 10 receptor, alpha


IL12A interleukin 12A (natural killer cell stimulatory factor 1, cytotoxic lymphocyte maturation factor 1, p35)

IL12B interleukin 12B (natural killer cell stimulatory factor 2, cytotoxic lymphocyte maturation factor 2, p40)

IL12RB1 interleukin 12 receptor, beta 1

IL13 interleukin 13

IL13RA1 interleukin 13 receptor, alpha 1

IL15 interleukin 15

IL16 interleukin 16

IL17A interleukin 17A

IL17B interleukin 17B

IL17F interleukin 17F

IL18 interleukin 18 (interferon-gamma-inducing factor)

IL18R1 interleukin 18 receptor 1

IL18RAP interleukin 18 receptor accessory protein

IL19 interleukin 19

IL1A interleukin 1, alpha

IL1B interleukin 1, beta

IL1R1 interleukin 1 receptor, type I

IL1R2 interleukin 1 receptor, type II

IL1RAP interleukin 1 receptor accessory protein

IL1RL1 interleukin 1 receptor-like 1

IL1RL2 interleukin 1 receptor-like 2

IL1RN interleukin 1 receptor antagonist

IL2 interleukin 2

IL20 interleukin 20

IL21 interleukin 21

IL21R interleukin 21 receptor

IL22 interleukin 22

IL22RA2 interleukin 22 receptor, alpha 2

IL23A interleukin 23, alpha subunit p19


IL26 interleukin 26

IL27 interleukin 27

IL28A interleukin 28A (interferon, lambda 2)

IL28A/B interleukin 28A (interferon, lambda 2)/interleukin 28B (interferon, lambda 3)

IL29 interleukin 29 (interferon, lambda 1)

306


IL2RB interleukin 2 receptor, beta

IL2RG interleukin 2 receptor, gamma

IL3 interleukin 3 (colony-stimulating factor, multiple)

IL32 interleukin 32

IL4 interleukin 4


IL5 interleukin 5 (colony-stimulating factor, eosinophil)

IL6 interleukin 6 (interferon, beta 2)


IL6ST interleukin 6 signal transducer (gp130, oncostatin M receptor)

IL7 interleukin 7


IL8 interleukin 8

IL9 interleukin 9

ILF3 interleukin enhancer binding factor 3, 90kDa

IRAK1 interleukin-1 receptor-associated kinase 1




IRF1 interferon regulatory factor 1






IRGM immunity-related GTPase family, M

ITGA2B integrin, alpha 2b (platelet glycoprotein IIb of IIb/IIIa complex, antigen CD41)

ITGA4 integrin, alpha 4 (antigen CD49D, alpha 4 subunit of VLA-4 receptor)

ITGA5 integrin, alpha 5 (fibronectin receptor, alpha polypeptide)

ITGA6 integrin, alpha 6

ITGAE integrin, alpha E (antigen CD103, human mucosal lymphocyte antigen 1; alpha polypeptide)

ITGAL integrin, alpha L (antigen CD11A (p180), lymphocyte function-associated antigen 1; alpha polypeptide)

ITGAM integrin, alpha M (complement component 3 receptor 3 subunit)

ITGAX integrin, alpha X (complement component 3 receptor 4 subunit)

ITGB1 integrin, beta 1 (fibronectin receptor, beta polypeptide, antigen CD29 includes MDF2, MSK12)

ITGB2 integrin, beta 2 (complement component 3 receptor 3 and 4 subunit)

ITLN1 intelectin 1 (galactofuranose binding)

ITLN2 intelectin 2

JAK1 Janus kinase 1

JAK2 Janus kinase 2

JAK3 Janus kinase 3

KCNJ2 potassium inwardly-rectifying channel, subfamily J, member 2

307

KIR_Activating_Subgroup_1 killer cell immunoglobulin-like receptor

KIR_Activating_Subgroup_2 killer cell immunoglobulin-like receptor

KIR_Inhibiting_Subgroup_1 killer cell immunoglobulin-like receptor

KIR_Inhibiting_Subgroup_2 killer cell immunoglobulin-like receptor

KIR3DL1 killer cell immunoglobulin-like receptor, three domains, long cytoplasmic tail, 1



KIT v-kit Hardy-Zuckerman 4 feline sarcoma viral oncogene homolog

KLRAP1 killer cell lectin-like receptor subfamily A pseudogene 1

KLRB1 killer cell lectin-like receptor subfamily B, member 1

KLRC1 killer cell lectin-like receptor subfamily C, member 1




KLRD1 killer cell lectin-like receptor subfamily D, member 1

KLRF1 killer cell lectin-like receptor subfamily F, member 1

KLRF2 killer cell lectin-like receptor subfamily F, member 2

KLRG1 killer cell lectin-like receptor subfamily G, member 1

KLRG2 killer cell lectin-like receptor subfamily G, member 2

KLRK1 killer cell lectin-like receptor subfamily K, member 1

LAG3 lymphocyte-activation gene 3

LAIR1 leukocyte-associated immunoglobulin-like receptor 1

LAMP3 lysosomal-associated membrane protein 3

LCK lymphocyte-specific protein tyrosine kinase

LCP2 lymphocyte cytosolic protein 2 (SH2 domain containing leukocyte protein of 76kDa)

LEF1 lymphoid enhancer-binding factor 1

LGALS3 lectin, galactoside-binding, soluble, 3

LIF leukemia inhibitory factor (cholinergic differentiation factor)

LILRA1 leukocyte immunoglobulin-like receptor, subfamily A (with TM domain), member 1


LILRA3 leukocyte immunoglobulin-like receptor, subfamily A (without TM domain), member 3




LILRB1 leukocyte immunoglobulin-like receptor, subfamily B (with TM and ITIM domains), member 1


308




LITAF lipopolysaccharide-induced TNF factor

LTA lymphotoxin alpha (TNF superfamily, member 1)

LTB4R leukotriene B4 receptor

LTB4R2 leukotriene B4 receptor 2

LTBR lymphotoxin beta receptor (TNFR superfamily, member 3)

LTF lactotransferrin

LY96 lymphocyte antigen 96

MAF v-maf musculoaponeurotic fibrosarcoma oncogene homolog (avian)

MALT1 mucosa associated lymphoid tissue lymphoma translocation gene 1

MAP4K1 mitogen-activated protein kinase kinase kinase kinase 1



MAPK1 mitogen-activated protein kinase 1



MAPKAPK2 mitogen-activated protein kinase-activated protein kinase 2

MARCO macrophage receptor with collagenous structure

MASP1 mannan-binding lectin serine peptidase 1 (C4/C2 activating component of Ra-reactive factor)

MASP2 mannan-binding lectin serine peptidase 2

MBL2 mannose-binding lectin (protein C) 2, soluble

MBP myelin basic protein

MCL1 myeloid cell leukemia sequence 1 (BCL2-related)

MIF macrophage migration inhibitory factor (glycosylation-inhibiting factor)

MME membrane metallo-endopeptidase

MR1 major histocompatibility complex, class I-related

MRC1 mannose receptor, C type 1

MS4A1 membrane-spanning 4-domains, subfamily A, member 1

MSR1 macrophage scavenger receptor 1

MUC1 mucin 1, cell surface associated

MX1 myxovirus (influenza virus) resistance 1, interferon-inducible protein p78 (mouse)

MYD88 myeloid differentiation primary response gene (88)

NCAM1 neural cell adhesion molecule 1

NCF4 neutrophil cytosolic factor 4, 40kDa

NCR1 natural cytotoxicity triggering receptor 1

NFATC1 nuclear factor of activated T-cells, cytoplasmic, calcineurin-dependent 1



NFIL3 nuclear factor, interleukin 3 regulated

309

NFKB1 nuclear factor of kappa light polypeptide gene enhancer in B-cells 1

NFKB2 nuclear factor of kappa light polypeptide gene enhancer in B-cells 2 (p49/p100)

NFKBIA nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, alpha

NFKBIZ nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, zeta

NLRP3 NLR family, pyrin domain containing 3

NOD1 nucleotide-binding oligomerization domain containing 1

NOD2 nucleotide-binding oligomerization domain containing 2

NOS2 nitric oxide synthase 2, inducible

NOTCH1 notch 1

NOTCH2 notch 2

NT5E 5'-nucleotidase, ecto (CD73)

PAX5 paired box 5

PDCD1 programmed cell death 1

PDCD1LG2 programmed cell death 1 ligand 2

PDCD2 programmed cell death 2

PDGFB platelet-derived growth factor beta polypeptide

PDGFRB platelet-derived growth factor receptor, beta polypeptide

PECAM1 platelet/endothelial cell adhesion molecule

PIGR polymeric immunoglobulin receptor

PLA2G2A phospholipase A2, group IIA (platelets, synovial fluid)

PLA2G2E phospholipase A2, group IIE

PLAU plasminogen activator, urokinase

PLAUR plasminogen activator, urokinase receptor

PML promyelocytic leukemia

POU2F2 POU class 2 homeobox 2

PPARG peroxisome proliferator-activated receptor gamma

PPBP pro-platelet basic protein (chemokine (C-X-C motif) ligand 7)

PRDM1 PR domain containing 1, with ZNF domain

PRF1 perforin 1 (pore forming protein)

PRKCD protein kinase C, delta

PSMB10 proteasome (prosome, macropain) subunit, beta type, 10



PSMB8 proteasome (prosome, macropain) subunit, beta type, 8 (large multifunctional peptidase 7)

PSMB9 proteasome (prosome, macropain) subunit, beta type, 9 (large multifunctional peptidase 2)

PSMC2 proteasome (prosome, macropain) 26S subunit, ATPase, 2

PSMD7 proteasome (prosome, macropain) 26S subunit, non-ATPase, 7

PTAFR platelet-activating factor receptor

PTGER4 prostaglandin E receptor 4 (subtype EP4)

PTGS2 prostaglandin-endoperoxide synthase 2 (prostaglandin G/H synthase and cyclooxygenase)

PTK2 PTK2 protein tyrosine kinase 2

PTPN2 protein tyrosine phosphatase, non-receptor type 2

PTPN22 protein tyrosine phosphatase, non-receptor type 22 (lymphoid)

310

PTPN6 protein tyrosine phosphatase, non-receptor type 6

PTPRC_all (common probe) protein tyrosine phosphatase, receptor type, C

CD45R0 protein tyrosine phosphatase, receptor type, C

CD45RA protein tyrosine phosphatase, receptor type, C

CD45RB protein tyrosine phosphatase, receptor type, C

PYCARD PYD and CARD domain containing

RAF1 v-raf-1 murine leukemia viral oncogene homolog 1

RAG1 recombination activating gene 1

RAG2 recombination activating gene 2

RARRES3 retinoic acid receptor responder (tazarotene induced) 3

RELA v-rel reticuloendotheliosis viral oncogene homolog A (avian)

RELB v-rel reticuloendotheliosis viral oncogene homolog B

RORC RAR-related orphan receptor C

RUNX1 runt-related transcription factor 1

S100A8 S100 calcium binding protein A8

S100A9 S100 calcium binding protein A9

S1PR1 sphingosine-1-phosphate receptor 1

SELE selectin E

SELL selectin L

SELPLG selectin P ligand

SERPING1 serpin peptidase inhibitor, clade G (C1 inhibitor), member 1

SH2D1A SH2 domain containing 1A

SIGIRR single immunoglobulin and toll-interleukin 1 receptor (TIR) domain

SKI v-ski sarcoma viral oncogene homolog (avian)

SLAMF1 signaling lymphocytic activation molecule family member 1

SLAMF6 SLAM family member 6

SLAMF7 SLAM family member 7

SLC2A1 solute carrier family 2 (facilitated glucose transporter), member 1

SMAD3 SMAD family member 3

SMAD5 SMAD family member 5

SOCS1 suppressor of cytokine signaling 1

SOCS3 suppressor of cytokine signaling 3

SPP1 secreted phosphoprotein 1

SRC v-src sarcoma (Schmidt-Ruppin A-2) viral oncogene homolog (avian)

STAT1 signal transducer and activator of transcription 1, 91kDa

STAT2 signal transducer and activator of transcription 2, 113kDa

STAT3 signal transducer and activator of transcription 3 (acute-phase response factor)

STAT4 signal transducer and activator of transcription 4

STAT5A signal transducer and activator of transcription 5A

STAT5B signal transducer and activator of transcription 5B

STAT6 signal transducer and activator of transcription 6, interleukin-4 induced

SYK spleen tyrosine kinase

TAGAP T-cell activation RhoGTPase activating protein

TAL1 T-cell acute lymphocytic leukemia 1

311

TAP1 transporter 1, ATP-binding cassette, sub-family B (MDR/TAP)

TAP2 transporter 2, ATP-binding cassette, sub-family B (MDR/TAP)

TAPBP TAP binding protein (tapasin)

TBK1 TANK-binding kinase 1

TBX21 T-box 21

TCF4 transcription factor 4

TCF7 transcription factor 7 (T-cell specific, HMG-box)

TFRC transferrin receptor (p90, CD71)

TGFB1 transforming growth factor, beta 1

TGFBI transforming growth factor, beta-induced, 68kDa

TGFBR1 transforming growth factor, beta receptor 1

TGFBR2 transforming growth factor, beta receptor II (70/80kDa)

THY1 Thy-1 cell surface antigen

TICAM1 toll-like receptor adaptor molecule 1

TIGIT T cell immunoreceptor with Ig and ITIM domains

TIRAP toll-interleukin 1 receptor (TIR) domain containing adaptor protein

TLR1 toll-like receptor 1








TMEM173 transmembrane protein 173

TNF tumor necrosis factor

TNFAIP3 tumor necrosis factor, alpha-induced protein 3

TNFAIP6 tumor necrosis factor, alpha-induced protein 6

TNFRSF10C tumor necrosis factor receptor superfamily, member 10c, decoy without an intracellular domain

TNFRSF11A tumor necrosis factor receptor superfamily, member 11a, NFKB activator

TNFRSF13B tumor necrosis factor receptor superfamily, member 13B

TNFRSF13C tumor necrosis factor receptor superfamily, member 13C

TNFRSF14 tumor necrosis factor receptor superfamily, member 14


TNFRSF1B tumor necrosis factor receptor superfamily, member 1B




TNFSF10 tumor necrosis factor (ligand) superfamily, member 10



TNFSF13B tumor necrosis factor (ligand) superfamily, member 13b




312

TOLLIP toll interacting protein

TP53 tumor protein p53

TRAF1 TNF receptor-associated factor 1






TYK2 tyrosine kinase 2

UBE2L3 ubiquitin-conjugating enzyme E2L 3

VCAM1 vascular cell adhesion molecule 1

VTN vitronectin

XBP1 X-box binding protein 1

XCL1 chemokine (C motif) ligand 1

XCR1 chemokine (C motif) receptor 1

ZAP70 zeta-chain (TCR) associated protein kinase 70kDa

ZBTB16 zinc finger and BTB domain containing 16

ZEB1 zinc finger E-box binding homeobox 1

Internal Reference Genes

ABCF1 ATP-binding cassette, sub-family F (GCN20), member 1

ALAS1 aminolevulinate, delta-, synthase 1

EEF1G eukaryotic translation elongation factor 1 gamma

G6PD glucose-6-phosphate dehydrogenase

GAPDH glyceraldehyde-3-phosphate dehydrogenase

GUSB glucuronidase, beta

HPRT1 hypoxanthine phosphoribosyltransferase 1

OAZ1 ornithine decarboxylase antizyme 1

POLR1B polymerase (RNA) I polypeptide B, 128kDa

POLR2A polymerase (RNA) II (DNA directed) polypeptide A, 220kDa

PPIA peptidylprolyl isomerase A (cyclophilin A)

SDHA succinate dehydrogenase complex, subunit A, flavoprotein (Fp)

TBP TATA box binding protein

TUBB tubulin, beta

RPL19 ribosomal protein L19 © 2013 NanoString

Technologies

microbiology and immunology of otitis media · otitis media chinh chung ngo master of science...

Documents