Burden of Disease and Change in Practice in Critically Ill
Transcript Of Burden of Disease and Change in Practice in Critically Ill
European Respiratory Journal
Burden of Disease and Change in Practice in Critically Ill Infants with Bronchiolitis
Journal: European Respiratory Journal
Manuscript ID ERJ-01648-2016.R1
Manuscript Type: Original Article
Date Submitted by the Author: 21-Jan-2017
Complete List of Authors:
Schlapbach, Luregn; University of Queensland, Paediatric Critical Care Research Group, Mater Research Institute Straney, Lahn; Monash University Gelbart, Ben; Royal Children's Hospital Melbourne Alexander, Jan; Lady Cilento Children's Hospital Franklin, Donna; University of Queensland, Paediatric Critical Care Research Group, Mater Research Institute Beca, John; Starship Children's Health Whitty, Jennifer; University of Queensland Ganu, Subodh; Royal Adelaide Hospital Wilkins, Barry; Children's Hospital at Westmead Slater, Anthony; Lady Cilento Children's Hospital Croston, Elizabeth; Princess Margaret Hospital for Children Erickson, Simon; Princess Margaret Hospital for Children Schibler, Andy; Paediatric Critical Care Research Group, Paediatric Intensive Care Unit
Key Words: bronchiolitis, infants, intensive care, ventilation, noninvasive ventilatory support
Page 1 of 45
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
European Respiratory Journal
Burden of Disease and Change in Practice in Critically Ill Infants with Bronchiolitis
Luregn J Schlapbach1,2,3*, Lahn Straney4*, Ben Gelbart5,6, Janet Alexander7,8, Donna Franklin1, John Beca9, Jennifer A. Whitty10,11, Subodh Ganu12,13, Barry Wilkins14, Anthony Slater2, Elizabeth Croston15, Simon Erickson15, Andreas Schibler1,2 on behalf of the Australian & New Zealand Intensive Care Society (ANZICS) Centre for Outcomes & Resource Evaluation (CORE) and the Australian & New Zealand Intensive Care Society (ANZICS) Paediatric Study Group
*LJ Schlapbach and L Straney contributed equally to this manuscript
1Paediatric Critical Care Research Group, Mater Research Institute, The University of Queensland, Brisbane, Australia 2Paediatric Intensive Care Unit, Lady Cilento Children’s Hospital, Brisbane, Australia 3Department of Pediatrics, Inselspital, Bern University Hospital, University of Bern, Switzerland 4Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia 5Paediatric Intensive Care Unit, The Royal Children’s Hospital, Melbourne, Australia 6Melbourne Children’s Research Institute, Melbourne Australia 7Australian and New Zealand Paediatric Intensive Care Registry, CORE, Lady Cilento Children’s Hospital Brisbane, Brisbane, Australia 8School of Medicine, The University of Queensland, Brisbane, Australia 9Paediatric Intensive Care Unit, Starship Children`s Hospital, Auckland, New Zealand 10School of Pharmacy, The University of Queensland, Brisbane, Australia
1
European Respiratory Journal
Page 2 of 45
1 2
11Health Economics Group, Norwich Medical School, University of East Anglia,
3
4
Norwich, United Kingdom
5
6
12University of Adelaide, Adelaide, Australia
7
8
13Paediatric Intensive Care Unit, Women’s and Children’s Hospital, Adelaide,
9
10
11
Australia
12
13
14Paediatric Intensive Care Unit, Children’s Hospital Westmead, Sydney, Australia
14
15
15Paediatric Intensive Care Unit, Princess Margaret Hospital for Children, Perth,
16
17
Australia
18
19
20
21
22
Correspondence:
23
24
Luregn J. Schlapbach, MD, FCICM
25
26
Paediatric Critical Care Research Group, Mater Research Institute, The University of
27
2289 Queensland
30
31
Paediatric Intensive Care Unit, Lady Cilento Children’s Hospital
32
33
South Brisbane QLD 4101 Australia
34
35
Phone
..61 (07) 3068 1111
36
37
email [email protected]
38
39
40
41
42
Running Title: Severe Bronchiolitis in Critically Ill Children
43
44
Word count text: 2990
45
46
47
4489 This article has an online data supplement
50
51
52
53
Take home message: Thresholds to admit bronchiolitis patients to PICU have
54
55
changed over the past decade with a major impact on costs and resource utilization.
56
57
58
59
60
2
Page 3 of 45
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
European Respiratory Journal
Authors and Contributions: Dr. Schlapbach and Dr. Straney had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Dr. Schlapbach was responsible for the study design, prepared the first manuscript draft and approved the final version. Drs Straney and Schibler were involved in study design, analyses, preparation of the first manuscript draft and final version. Dr. Slater led the design and maintenance of the ANZPIC Registry since inception, was involved in study design, and approved the final version. Dr. Whitty and Mrs Franklin performed healthcare cost related analyses and interpretation. Drs Beca, Wilkins, Erickson, Croston, Gebhard, Ganu, and Mrs Alexander were involved in study design, data collection, and manuscript preparation, and approved the final version.
3
European Respiratory Journal
Page 4 of 45
1
2
ABSTRACT
3
4
Bronchiolitis represents the most common cause for non-elective admission to
5
6
pediatric intensive care units (ICUs).
7
8
We assessed changes in admission rate, respiratory support, and outcomes of
9
10
11
infants <24 months with bronchiolitis admitted to ICU between 2002 and 2014 in
12
13
Australia and New Zealand.
14
15
During the study period, bronchiolitis was responsible for 9,628 (27.6%) of 34,829
16
17
non-elective ICU admissions. The estimated population-based ICU admission rate
18
19
20
due to bronchiolitis increased by 11.76/100,000 each year (95%-CI 8.11-15.41). The
21
22
proportion of bronchiolitis patients requiring intubation decreased from 36.8% in
23
24
2002, to 10.8% in 2014 (adjusted OR 0.35; 0.27-0.46), whilst a dramatic increase in
25
26
high-flow nasal cannula therapy use to 72.6% was observed (p<0.001). We observed
27
2289 a considerable variability in practice between units, with six-fold differences in risk-
30
31
adjusted intubation rates which were not explained by ICU type, size, nor major
32
33
patient factors. Annual direct hospitalization costs due to severe bronchiolitis
34
35
increased to over US$30 M in 2014.
36
37
In conclusion, we observed increasing health care burden due to severe bronchiolitis,
38
39
40
with a major change in practice in the management from invasive to non-invasive
41
42
support suggesting thresholds to admit bronchiolitis patients to ICU have changed.
43
44
Future studies should assess strategies for management of bronchiolitis outside
45
46
ICUs.
47
48
49
50
51
Abstract Word count: 199
52
53
.
54
55
56
57
58
59
60
4
Page 5 of 45
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
European Respiratory Journal
Keywords: infant; ventilation; bronchiolitis; intensive care; high-flow nasal cannulae
Abbreviations:
OR
Odd’s ratio
PICU
Paediatric Intensive Care Unit
PIM
Paediatric Index of Mortality
IV
Invasive ventilation
NIV
non-invasive ventilation
CPAP
continuous positive airway pressure
BIPAP
biphasic positive airway pressure
HFNC
high-flow nasal cannulae
5
European Respiratory Journal
Page 6 of 45
1
2
Introduction
3
4
5
6
Bronchiolitis is a common viral lower respiratory tract infection in infants
7
8
characterized by acute small airway inflammation, and represents the leading cause
9
10
11
for hospital admission during the first year of life[1]. In high income countries,
12
13
approximately one out of eight infants hospitalized with bronchiolitis require
14
15
admission to Intensive Care Units (ICU) for respiratory support as a result of
16
17
progressive respiratory distress with respiratory failure and hypoxemia[2, 3]. Despite
18
19
20
a general trend towards a reduction in hospital admissions overall, bronchiolitis-
21
22
related hospitalization costs have recently increased amounting to US$1.73 billion
23
24
per year in the US[4]. In the past decades, pharmacological interventions have failed
25
26
to show any benefit, and as a result, consensus guidelines emphasize supportive
27
2289 treatment options[2, 3, 5-8]. While invasive ventilation (IV) was traditionally
30
31
considered the cornerstone of treatment for severe bronchiolitis in ICU, over recent
32
33
years, an increasing number of single center studies have reported benefits
34
35
associated with early use of non-invasive ventilation (NIV) and High-Flow Nasal
36
37
Cannula (HFNC) therapy to reduce the need for intubation and IV in bronchiolitis[9-
38
39
40
13].
41
42
43
44
The aims of this study were to describe the population-based admission rate and
45
46
severity of bronchiolitis in infants in Australia and New Zealand admitted to intensive
47
4489 care, to determine risk factors for invasive ventilation, and to assess trends in
50
51
admission rate, management, outcome and associated direct health care costs over
52
53
a 13 year period from 2002-2014.
54
55
56
57
58
59
60
6
Page 7 of 45
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
European Respiratory Journal
Methods (see as well Online Supplementary Material)
A multicenter, binational, retrospective study of all patients reported to the Australian and New Zealand Paediatric Intensive Care (ANZPIC) Registry[14]. The study was approved by the Human Research and Ethics Committee (Mater Health Services HREC, Brisbane, Australia) including waiver of informed consent. The ANZPIC Registry prospectively records demographics, physiologic variables at admission, intensive care support, diagnoses and outcomes of PICU and general ICU admissions in children <16 years of age in Australia and New Zealand[14], and captures 92-94% of all pediatric ICU admissions. Inclusion and Exclusion Criteria: Infants aged <729 days and admitted with a diagnosis of bronchiolitis[8] and who were admitted to a pediatric ICU (PICU) or a general ICU in Australia or New Zealand between 1st January 2002 and 31st December 2014 were included. Elective admissions were excluded and infants with preexisting tracheostomies were excluded. Outcomes and definitions: The primary outcome was defined as the proportion of infants requiring intubation and invasive ventilation (IV). Non-invasive ventilation (NIV) was defined as continous positive airway pressure (CPAP) with or without pressure support delivered through a nasal mask, full-face mask, or a nasopharyngeal tube. Mechanical ventilation was defined as either IV and/or NIV. Since 2010, the ANZPIC registry has been prospectively recording the use of highflow nasal cannulae oxygen (HFNC) therapy. HFNC was defined as > 1L/kg/min flow of a gas oxygen mixture through nasal cannula, and was coded separately from mechanical ventilation support [12, 13]. Data analyses were therefore separated into two periods: a period before widespread use of HFNC therapy (2002-2009) and a period post widespread introduction of HFNC therapy (2010-2014). With the
7
European Respiratory Journal
Page 8 of 45
1
2
exception of one PICU, HFNC was not routinely used in the main PICUs and ICUs
3
4
prior to 2010.
5
6
Cost estimates methodology are provided in the Supplementary Material.
7
8
9
10
11
Statistics:
12
13
Data are presented as percentages and numbers or means with standard deviation.
14
15
T-tests were used to compare subgroups. Population-based admission rate
16
17
estimates were calculated. We assessed linear trends in respiratory support over the
18
19
20
13-year period. In addition, trends during the 13-year study period were assessed by
21
22
comparing risk-adjusted need for invasive ventilation. We constructed a multivariate
23
24
prediction model for the need for invasive ventilation. For multivariable models, all
25
26
significant predictors from the univariable analyses were used. We used a backward
27
2289 stepwise elimination procedure to eliminate non-significant predictors based on
30
31
p>0.05.
32
33
All PICUs in Australia and New Zealand contributed to the ANZPIC registry for the
34
35
entire duration of the study period. The number of general ICUs contributing to the
36
37
registry increased from 6 to 19 during the study period. In order to account for
38
39
40
potential reporting bias, the following predefined subgroup analyses were performed:
41
42
i) specialized pediatric ICUs (PICUs); ii) general (mixed adult and pediatric) ICUs; iii)
43
44
pediatric and general ICUs that had been contributing for the entire length of the
45
46
study period to the ANZPIC registry.
47
4489 Further details of risk prediction models are provided in the Supplementary Material.
50
51
All analyses were conducted using Stata (version 12.1, Stata Corp, College Station,
52
53
Texas, USA). P-values less than 0.05 were considered significant.
54
55
56
57
58
59
60
8
Page 9 of 45
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
European Respiratory Journal
Results
During the study period, Bronchiolitis was the most common cause of ICU admission, and was responsible for 9,628 (27.6%) of 34,829 non-elective admissions in infants below two years of age. 324 infants with tracheostomies in situ at time of admission were excluded. Prematurity (20 %), chronic respiratory conditions (10%) and congenital cardiac disease (7 %) were the most common underlying conditions (Table 1). During 2010 to 2014 (post widespread introduction of HFNC), 5670 infants were admitted with bronchiolitis in comparison to 3634 during 2002 to 2009 (Table 2). In recent years, infants with bronchiolitis admitted to ICU were older, less likely to retrieved, more likely to be admitted to a general ICU, and less likely to have underlying diseases (p<0.001). The average severity of disease as measured by PIM2 decreased significantly, and ICU and hospital LOS decreased accordingly. The crude mortality over the entire study period was 0.38% (35/9304). The re-calibrated PIM2 standardized mortality ratio declined from 1.53 (0.99-2.26) in 2002-2009 down to 0.54 (0.26-0.98) in 2010-2014.
The annual number of infants with bronchiolitis admitted to ICU (including pediatric and general ICUs) increased from 383 cases in 2002 to 1528 cases in 2014 (Table 2, Figure 1). The total number of all non-elective ICU admissions per year during this time increased from 1933 to 4115. The estimated population-based age-standardised admission rate of bronchiolitis increased during the study period with an average annual increase of 11.76/100,000 infants <24 months (95%-CI 8.11 to 15.41). The change in admission rate was most pronounced in general ICUs, which took 40% of all bronchiolitis admissions requiring intensive care in 2014 (p<0.0001). The increase in admission was less marked when restricting analyses to PICUs (annual increase
9
Burden of Disease and Change in Practice in Critically Ill Infants with Bronchiolitis
Journal: European Respiratory Journal
Manuscript ID ERJ-01648-2016.R1
Manuscript Type: Original Article
Date Submitted by the Author: 21-Jan-2017
Complete List of Authors:
Schlapbach, Luregn; University of Queensland, Paediatric Critical Care Research Group, Mater Research Institute Straney, Lahn; Monash University Gelbart, Ben; Royal Children's Hospital Melbourne Alexander, Jan; Lady Cilento Children's Hospital Franklin, Donna; University of Queensland, Paediatric Critical Care Research Group, Mater Research Institute Beca, John; Starship Children's Health Whitty, Jennifer; University of Queensland Ganu, Subodh; Royal Adelaide Hospital Wilkins, Barry; Children's Hospital at Westmead Slater, Anthony; Lady Cilento Children's Hospital Croston, Elizabeth; Princess Margaret Hospital for Children Erickson, Simon; Princess Margaret Hospital for Children Schibler, Andy; Paediatric Critical Care Research Group, Paediatric Intensive Care Unit
Key Words: bronchiolitis, infants, intensive care, ventilation, noninvasive ventilatory support
Page 1 of 45
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
European Respiratory Journal
Burden of Disease and Change in Practice in Critically Ill Infants with Bronchiolitis
Luregn J Schlapbach1,2,3*, Lahn Straney4*, Ben Gelbart5,6, Janet Alexander7,8, Donna Franklin1, John Beca9, Jennifer A. Whitty10,11, Subodh Ganu12,13, Barry Wilkins14, Anthony Slater2, Elizabeth Croston15, Simon Erickson15, Andreas Schibler1,2 on behalf of the Australian & New Zealand Intensive Care Society (ANZICS) Centre for Outcomes & Resource Evaluation (CORE) and the Australian & New Zealand Intensive Care Society (ANZICS) Paediatric Study Group
*LJ Schlapbach and L Straney contributed equally to this manuscript
1Paediatric Critical Care Research Group, Mater Research Institute, The University of Queensland, Brisbane, Australia 2Paediatric Intensive Care Unit, Lady Cilento Children’s Hospital, Brisbane, Australia 3Department of Pediatrics, Inselspital, Bern University Hospital, University of Bern, Switzerland 4Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia 5Paediatric Intensive Care Unit, The Royal Children’s Hospital, Melbourne, Australia 6Melbourne Children’s Research Institute, Melbourne Australia 7Australian and New Zealand Paediatric Intensive Care Registry, CORE, Lady Cilento Children’s Hospital Brisbane, Brisbane, Australia 8School of Medicine, The University of Queensland, Brisbane, Australia 9Paediatric Intensive Care Unit, Starship Children`s Hospital, Auckland, New Zealand 10School of Pharmacy, The University of Queensland, Brisbane, Australia
1
European Respiratory Journal
Page 2 of 45
1 2
11Health Economics Group, Norwich Medical School, University of East Anglia,
3
4
Norwich, United Kingdom
5
6
12University of Adelaide, Adelaide, Australia
7
8
13Paediatric Intensive Care Unit, Women’s and Children’s Hospital, Adelaide,
9
10
11
Australia
12
13
14Paediatric Intensive Care Unit, Children’s Hospital Westmead, Sydney, Australia
14
15
15Paediatric Intensive Care Unit, Princess Margaret Hospital for Children, Perth,
16
17
Australia
18
19
20
21
22
Correspondence:
23
24
Luregn J. Schlapbach, MD, FCICM
25
26
Paediatric Critical Care Research Group, Mater Research Institute, The University of
27
2289 Queensland
30
31
Paediatric Intensive Care Unit, Lady Cilento Children’s Hospital
32
33
South Brisbane QLD 4101 Australia
34
35
Phone
..61 (07) 3068 1111
36
37
email [email protected]
38
39
40
41
42
Running Title: Severe Bronchiolitis in Critically Ill Children
43
44
Word count text: 2990
45
46
47
4489 This article has an online data supplement
50
51
52
53
Take home message: Thresholds to admit bronchiolitis patients to PICU have
54
55
changed over the past decade with a major impact on costs and resource utilization.
56
57
58
59
60
2
Page 3 of 45
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
European Respiratory Journal
Authors and Contributions: Dr. Schlapbach and Dr. Straney had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Dr. Schlapbach was responsible for the study design, prepared the first manuscript draft and approved the final version. Drs Straney and Schibler were involved in study design, analyses, preparation of the first manuscript draft and final version. Dr. Slater led the design and maintenance of the ANZPIC Registry since inception, was involved in study design, and approved the final version. Dr. Whitty and Mrs Franklin performed healthcare cost related analyses and interpretation. Drs Beca, Wilkins, Erickson, Croston, Gebhard, Ganu, and Mrs Alexander were involved in study design, data collection, and manuscript preparation, and approved the final version.
3
European Respiratory Journal
Page 4 of 45
1
2
ABSTRACT
3
4
Bronchiolitis represents the most common cause for non-elective admission to
5
6
pediatric intensive care units (ICUs).
7
8
We assessed changes in admission rate, respiratory support, and outcomes of
9
10
11
infants <24 months with bronchiolitis admitted to ICU between 2002 and 2014 in
12
13
Australia and New Zealand.
14
15
During the study period, bronchiolitis was responsible for 9,628 (27.6%) of 34,829
16
17
non-elective ICU admissions. The estimated population-based ICU admission rate
18
19
20
due to bronchiolitis increased by 11.76/100,000 each year (95%-CI 8.11-15.41). The
21
22
proportion of bronchiolitis patients requiring intubation decreased from 36.8% in
23
24
2002, to 10.8% in 2014 (adjusted OR 0.35; 0.27-0.46), whilst a dramatic increase in
25
26
high-flow nasal cannula therapy use to 72.6% was observed (p<0.001). We observed
27
2289 a considerable variability in practice between units, with six-fold differences in risk-
30
31
adjusted intubation rates which were not explained by ICU type, size, nor major
32
33
patient factors. Annual direct hospitalization costs due to severe bronchiolitis
34
35
increased to over US$30 M in 2014.
36
37
In conclusion, we observed increasing health care burden due to severe bronchiolitis,
38
39
40
with a major change in practice in the management from invasive to non-invasive
41
42
support suggesting thresholds to admit bronchiolitis patients to ICU have changed.
43
44
Future studies should assess strategies for management of bronchiolitis outside
45
46
ICUs.
47
48
49
50
51
Abstract Word count: 199
52
53
.
54
55
56
57
58
59
60
4
Page 5 of 45
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
European Respiratory Journal
Keywords: infant; ventilation; bronchiolitis; intensive care; high-flow nasal cannulae
Abbreviations:
OR
Odd’s ratio
PICU
Paediatric Intensive Care Unit
PIM
Paediatric Index of Mortality
IV
Invasive ventilation
NIV
non-invasive ventilation
CPAP
continuous positive airway pressure
BIPAP
biphasic positive airway pressure
HFNC
high-flow nasal cannulae
5
European Respiratory Journal
Page 6 of 45
1
2
Introduction
3
4
5
6
Bronchiolitis is a common viral lower respiratory tract infection in infants
7
8
characterized by acute small airway inflammation, and represents the leading cause
9
10
11
for hospital admission during the first year of life[1]. In high income countries,
12
13
approximately one out of eight infants hospitalized with bronchiolitis require
14
15
admission to Intensive Care Units (ICU) for respiratory support as a result of
16
17
progressive respiratory distress with respiratory failure and hypoxemia[2, 3]. Despite
18
19
20
a general trend towards a reduction in hospital admissions overall, bronchiolitis-
21
22
related hospitalization costs have recently increased amounting to US$1.73 billion
23
24
per year in the US[4]. In the past decades, pharmacological interventions have failed
25
26
to show any benefit, and as a result, consensus guidelines emphasize supportive
27
2289 treatment options[2, 3, 5-8]. While invasive ventilation (IV) was traditionally
30
31
considered the cornerstone of treatment for severe bronchiolitis in ICU, over recent
32
33
years, an increasing number of single center studies have reported benefits
34
35
associated with early use of non-invasive ventilation (NIV) and High-Flow Nasal
36
37
Cannula (HFNC) therapy to reduce the need for intubation and IV in bronchiolitis[9-
38
39
40
13].
41
42
43
44
The aims of this study were to describe the population-based admission rate and
45
46
severity of bronchiolitis in infants in Australia and New Zealand admitted to intensive
47
4489 care, to determine risk factors for invasive ventilation, and to assess trends in
50
51
admission rate, management, outcome and associated direct health care costs over
52
53
a 13 year period from 2002-2014.
54
55
56
57
58
59
60
6
Page 7 of 45
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
European Respiratory Journal
Methods (see as well Online Supplementary Material)
A multicenter, binational, retrospective study of all patients reported to the Australian and New Zealand Paediatric Intensive Care (ANZPIC) Registry[14]. The study was approved by the Human Research and Ethics Committee (Mater Health Services HREC, Brisbane, Australia) including waiver of informed consent. The ANZPIC Registry prospectively records demographics, physiologic variables at admission, intensive care support, diagnoses and outcomes of PICU and general ICU admissions in children <16 years of age in Australia and New Zealand[14], and captures 92-94% of all pediatric ICU admissions. Inclusion and Exclusion Criteria: Infants aged <729 days and admitted with a diagnosis of bronchiolitis[8] and who were admitted to a pediatric ICU (PICU) or a general ICU in Australia or New Zealand between 1st January 2002 and 31st December 2014 were included. Elective admissions were excluded and infants with preexisting tracheostomies were excluded. Outcomes and definitions: The primary outcome was defined as the proportion of infants requiring intubation and invasive ventilation (IV). Non-invasive ventilation (NIV) was defined as continous positive airway pressure (CPAP) with or without pressure support delivered through a nasal mask, full-face mask, or a nasopharyngeal tube. Mechanical ventilation was defined as either IV and/or NIV. Since 2010, the ANZPIC registry has been prospectively recording the use of highflow nasal cannulae oxygen (HFNC) therapy. HFNC was defined as > 1L/kg/min flow of a gas oxygen mixture through nasal cannula, and was coded separately from mechanical ventilation support [12, 13]. Data analyses were therefore separated into two periods: a period before widespread use of HFNC therapy (2002-2009) and a period post widespread introduction of HFNC therapy (2010-2014). With the
7
European Respiratory Journal
Page 8 of 45
1
2
exception of one PICU, HFNC was not routinely used in the main PICUs and ICUs
3
4
prior to 2010.
5
6
Cost estimates methodology are provided in the Supplementary Material.
7
8
9
10
11
Statistics:
12
13
Data are presented as percentages and numbers or means with standard deviation.
14
15
T-tests were used to compare subgroups. Population-based admission rate
16
17
estimates were calculated. We assessed linear trends in respiratory support over the
18
19
20
13-year period. In addition, trends during the 13-year study period were assessed by
21
22
comparing risk-adjusted need for invasive ventilation. We constructed a multivariate
23
24
prediction model for the need for invasive ventilation. For multivariable models, all
25
26
significant predictors from the univariable analyses were used. We used a backward
27
2289 stepwise elimination procedure to eliminate non-significant predictors based on
30
31
p>0.05.
32
33
All PICUs in Australia and New Zealand contributed to the ANZPIC registry for the
34
35
entire duration of the study period. The number of general ICUs contributing to the
36
37
registry increased from 6 to 19 during the study period. In order to account for
38
39
40
potential reporting bias, the following predefined subgroup analyses were performed:
41
42
i) specialized pediatric ICUs (PICUs); ii) general (mixed adult and pediatric) ICUs; iii)
43
44
pediatric and general ICUs that had been contributing for the entire length of the
45
46
study period to the ANZPIC registry.
47
4489 Further details of risk prediction models are provided in the Supplementary Material.
50
51
All analyses were conducted using Stata (version 12.1, Stata Corp, College Station,
52
53
Texas, USA). P-values less than 0.05 were considered significant.
54
55
56
57
58
59
60
8
Page 9 of 45
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
European Respiratory Journal
Results
During the study period, Bronchiolitis was the most common cause of ICU admission, and was responsible for 9,628 (27.6%) of 34,829 non-elective admissions in infants below two years of age. 324 infants with tracheostomies in situ at time of admission were excluded. Prematurity (20 %), chronic respiratory conditions (10%) and congenital cardiac disease (7 %) were the most common underlying conditions (Table 1). During 2010 to 2014 (post widespread introduction of HFNC), 5670 infants were admitted with bronchiolitis in comparison to 3634 during 2002 to 2009 (Table 2). In recent years, infants with bronchiolitis admitted to ICU were older, less likely to retrieved, more likely to be admitted to a general ICU, and less likely to have underlying diseases (p<0.001). The average severity of disease as measured by PIM2 decreased significantly, and ICU and hospital LOS decreased accordingly. The crude mortality over the entire study period was 0.38% (35/9304). The re-calibrated PIM2 standardized mortality ratio declined from 1.53 (0.99-2.26) in 2002-2009 down to 0.54 (0.26-0.98) in 2010-2014.
The annual number of infants with bronchiolitis admitted to ICU (including pediatric and general ICUs) increased from 383 cases in 2002 to 1528 cases in 2014 (Table 2, Figure 1). The total number of all non-elective ICU admissions per year during this time increased from 1933 to 4115. The estimated population-based age-standardised admission rate of bronchiolitis increased during the study period with an average annual increase of 11.76/100,000 infants <24 months (95%-CI 8.11 to 15.41). The change in admission rate was most pronounced in general ICUs, which took 40% of all bronchiolitis admissions requiring intensive care in 2014 (p<0.0001). The increase in admission was less marked when restricting analyses to PICUs (annual increase
9