Pediatric Pulmonology
Childhood Interstitial Lung Disease: A Systematic Review Neil J Hime, PhD,1,2 Yvonne Zurynski, PhD,1,2 Dominic Fitzgerald, PhD,3,4 Hiran Selvadurai, PhD,3,4 Amy Phu, MScMed (ClinEpid),1,2 Marie Deverell, PhD,1,2 Elizabeth J Elliott, MD,1,2,5 and Adam Jaffe, MD6,7 Summary. Objectives: Childhood interstitial lung disease (chILD) is a group of rare chronic and complex disorders of variable pathology. There has been no systematic review of published chILD research. This study aimed to describe chILD classification systems, epidemiology, morbidity, treatments, outcomes, and the impact of chILD on families and the burden on health services. Methods: A systematic literature search for original studies on chILD was undertaken in the major biomedical databases to the end of December 2013. Epidemiological studies, case series and studies describing classification systems were included. Single case studies were excluded. Results: The search yielded 37 publications that met study criteria. Four different chILD classification systems have been proposed in the past decade. The incidence of chILD has been estimated at 0.13–16.2 cases/100,000 children/year. One to five new cases presented to individual hospitals each year. In developed countries, the median mortality was 13% (6–19%). Morbidity and outcomes were highly variable and not systematically reported. Corticosteroids and hydroxychloroquine were the most common treatments. The impact of chILD on families and the burden on health services has not been studied. Conclusions: The heterogeneity of the chILD group of disorders, different determinations of what constitutes a chILD disorder and, a paucity of large epidemiological studies precludes consolidation of results across studies. Consensus on chILD classification is needed to support diagnosis and allow direct comparisons of research evidence. Active disease surveillance and international patient registries are required to advance understanding and management of chILD. Pediatr Pulmonol. ß 2015 Wiley Periodicals, Inc.
Key words: Childhood interstitial lung disease; chILD syndrome; Interstitial lung disease; Diffuse lung disease. Funding source: Australian Government Department of Health, Chronic Disease Prevention and Service Improvement Fund., National Health and Medical Research Council of Australia (Number 1021480).
INTRODUCTION
Childhood interstitial lung disease (chILD) describes a heterogeneous group of rare chronic respiratory disorders in children, particularly in infants under 2 years of age, which includes growth and developmental abnormalities
as well as immunological problems. These disorders have variable lung pathology associated with poor gas exchange.1–4 chILD is clinically complex and associated with high morbidity and mortality.4,5 The typical features
1
Discipline of Paediatrics and Child Health, Sydney Medical School, The University of Sydney, Sydney, Australia.
7 Discipline of Paediatrics, School of Women’s and Children’s Health, UNSW Medicine, The University of New South Wales, Sydney, Australia.
2 Australian Paediatric Surveillance Unit, Kids Research Institute, Westmead NSW 2145, Australia.
Conflict of interest: None
3
Department of Respiratory Medicine, The Children’s Hospital at Westmead, Westmead NSW 2145, Australia. 4 Discipline of Paediatrics and Child Health, Sydney Medical School, The University of Sydney Clinical School, Sydney, Australia. 5
Correspondence to: Professor Adam Jaffe, School of Women’s and Children’s Health, Medicine UNSW, University of New South Wales, c/o Sydney Children’s Hospital, High Street, Randwick, New South Wales 2031, Australia E: [email protected] Received 30 June 2014; Revised 23 December 2014; Accepted 15 January 2015.
The Children’s Hospital at Westmead, Sydney, Australia.
6
Department of Respiratory Medicine, Sydney Children’s Hospital, Randwick NSW 2031, Australia.
ß 2015 Wiley Periodicals, Inc.
DOI 10.1002/ppul.23183 Published online in Wiley Online Library (wileyonlinelibrary.com).
2
Hime et al.
of chILD are breathlessness and hypoxemia in the presence of diffuse lung infiltrates on radiography or computed tomography. The term “diffuse lung disease” has often been used interchangeably, because the interstitial compartment is not always involved.1,2 Recently, the term “chILD syndrome” has been adopted by the American Thoracic Society Committee on chILD and the chILD Research Network in order to provide a set of criteria to aid the diagnosis of uncommon causes of diffuse lung disease phenotypes in infants.1 chILD differs from ILD in adults in that it is much rarer and, includes pathologies that are unique to infants and young children (e.g., pulmonary interstitial glycogenosis (PIG),6 neuroendocrine cell hyperplasia of infancy (NEHI)7, and chronic pneumonitis of infancy,8 and genetic abnormalities that cause ILD in childhood (e.g., surfactant protein deficiencies9). Conversely, common adult ILD such as usual interstitial pneumonia (UIP) is rarely described in children. Because individual disorders within the chILD group are extremely rare, this hinders research into this important area.10 Given the chronicity and high morbidity of chILD, the impact on families and burden on health services are likely to be significant. In this systematic literature review we aim to describe:
classification systems that have been applied to chILD the frequency of chILD in populations and the frequency of incident cases in hospitals
the morbidity, treatment, and outcomes of chILD the impact of chILD on families and the burden on health services
METHODS Identification of Relevant Studies
The following databases were searched: Medline (1946–2013), Embase (1966–2013), Evidence-Based Medicine (EBM) Reviews—Cochrane Database of Systematic Reviews (2005–2013), EBM Reviews— Cochrane Central Register of Controlled Trials (1991– 2013), Informit—Australasian Medical Index (1968– 2009), Informit—Health Collection (1977–2013), and clinicaltrials.gov (2000–2013). The following Medical Subject Headings (MeSH) were used in search strategies: lung diseases, interstitial; incidence; prevalence; public health; economics, medical; cost-benefit analysis; health services; health care rationing; health care costs. Forty text words (supplementary material, S1) were applied to searches independent of MeSH terms. The methodological filters “all child (0–18 years)” and “humans” were applied to all searches in Medline. NH and AP identified potentially relevant studies through screening of titles and abstracts. NH Pediatric Pulmonology
examined the full texts of potentially relevant studies and studies that fit the inclusion/exclusion criteria were included for review. To identify additional studies for inclusion, NH and AP screened the reference lists of review articles and relevant studies. Only articles in English were retrieved. Study Inclusion Criteria
Studies reporting original data in children aged 0–18 years, on the following aspects of chILD were included: classification systems, frequency (population prevalence/ incidence or hospital-based case series), symptoms, morbidities, treatments, outcomes, impacts on families, burdens on health services. Because chILD encompasses a heterogeneous group of disorders, this review is not limited to studies that included all chILD disorders but also includes studies that investigated specific chILD disorders. While we have used the currently accepted terminology of “chILD” in this systematic review, many of the epidemiological, and case series studies included for review were conducted prior to the development of the recent proposed classification schemes. Study Exclusion Criteria
Case studies reporting single patients or family groups were excluded. Studies involving adults (over 18 years of age) were excluded unless they reported sub-analyses of children. Studies reporting diagnostic evaluation without reporting on morbidity, treatment or outcome were excluded. Studies with insufficient detail to determine how and in whom the research was conducted were excluded. Also excluded from this review were common causes of diffuse lung disease that are not encompassed within “chILD syndrome”. As stated in the current clinical practice guidelines, these common causes include: cystic fibrosis, congenital, or acquired immunodeficiency, congenital heart disease, bronchopulmonary dysplasia, pulmonary infection, primary ciliary dyskinesia presenting with newborn respiratory distress, and recurrent aspiration.1 Data Extraction
NH and AP extracted data. Where possible, units of frequency were adjusted to facilitate comparisons between studies. RESULTS Literature search
A total of 37 studies of chILD disorders were included (Fig. 1).
4 presented classification systems for chILD 4 estimated population prevalence or incidence
Childhood Interstitial Lung Disease
3
included a group of disorders with onset of disease in infancy. Chronic aspiration syndromes, resolving acute respiratory distress syndrome and bronchopulmonary dysplasia were excluded. In the most recent chILD classification system, proposed by the American Thoracic Society Committee on chILD and the chILD Research Network, the term “chILD syndrome” was used to exclude common causes of diffuse lung disease such as cystic fibrosis, congenital heart disease, bronchopulmonary dysplasia, and pulmonary infection; it recognised that some chILD conditions may be asymptomatic when identified.1 It was acknowledged that many of the diffuse lung diseases (proposed by Deutsch et al.2) may present as chILD syndrome. Frequency of chILD in Populations and Hospitals
Fig. 1. Results of literature search strategy.
5 reported hospital-based incident cases of chILD within a specific time period
33 described morbidities, treatments, and/or outcomes Some studies reported data applicable to more than one of the categories above. We found no studies that reported impacts on families or burdens on health services. Classification Systems Applied to chILD
Four studies published between 2004 and 2013 describe classification systems for chILD (Table 1). There are inconsistencies in the definitions of chILD, each classification system included different disorders. Based on 185 cases of ILD in immunocompetent children in the 0–18 year age group, the European Respiratory Society Taskforce11 defined sub-groups according to aetiology and histopathology, modifying a classification of interstitial pneumonias in adults. Deutsch et al.2 classified 187 cases of diffuse lung disease using clinical data, imaging, and lung biopsy in children aged less than two years. They defined “diffuse lung disease” and included a broad range of diseases, some of which primarily affect distal airways outside of the interstitium. The classification had an emphasis on pathological findings and, disorders considered more prevalent in infancy were distinguished from other diffuse lung diseases. In 2010 Clement et al.5 proposed a step-wise approach to diagnosis of chILD in which specific disorders were grouped by aetiology. This classification
Three studies12–14 reported estimates of the frequency of chILD in populations (Table 2). Each study used different methods of case ascertainment and case definitions. Incidence estimates varied from 0.13 cases/ 100,000 children
Childhood Interstitial Lung Disease: A Systematic Review Neil J Hime, PhD,1,2 Yvonne Zurynski, PhD,1,2 Dominic Fitzgerald, PhD,3,4 Hiran Selvadurai, PhD,3,4 Amy Phu, MScMed (ClinEpid),1,2 Marie Deverell, PhD,1,2 Elizabeth J Elliott, MD,1,2,5 and Adam Jaffe, MD6,7 Summary. Objectives: Childhood interstitial lung disease (chILD) is a group of rare chronic and complex disorders of variable pathology. There has been no systematic review of published chILD research. This study aimed to describe chILD classification systems, epidemiology, morbidity, treatments, outcomes, and the impact of chILD on families and the burden on health services. Methods: A systematic literature search for original studies on chILD was undertaken in the major biomedical databases to the end of December 2013. Epidemiological studies, case series and studies describing classification systems were included. Single case studies were excluded. Results: The search yielded 37 publications that met study criteria. Four different chILD classification systems have been proposed in the past decade. The incidence of chILD has been estimated at 0.13–16.2 cases/100,000 children/year. One to five new cases presented to individual hospitals each year. In developed countries, the median mortality was 13% (6–19%). Morbidity and outcomes were highly variable and not systematically reported. Corticosteroids and hydroxychloroquine were the most common treatments. The impact of chILD on families and the burden on health services has not been studied. Conclusions: The heterogeneity of the chILD group of disorders, different determinations of what constitutes a chILD disorder and, a paucity of large epidemiological studies precludes consolidation of results across studies. Consensus on chILD classification is needed to support diagnosis and allow direct comparisons of research evidence. Active disease surveillance and international patient registries are required to advance understanding and management of chILD. Pediatr Pulmonol. ß 2015 Wiley Periodicals, Inc.
Key words: Childhood interstitial lung disease; chILD syndrome; Interstitial lung disease; Diffuse lung disease. Funding source: Australian Government Department of Health, Chronic Disease Prevention and Service Improvement Fund., National Health and Medical Research Council of Australia (Number 1021480).
INTRODUCTION
Childhood interstitial lung disease (chILD) describes a heterogeneous group of rare chronic respiratory disorders in children, particularly in infants under 2 years of age, which includes growth and developmental abnormalities
as well as immunological problems. These disorders have variable lung pathology associated with poor gas exchange.1–4 chILD is clinically complex and associated with high morbidity and mortality.4,5 The typical features
1
Discipline of Paediatrics and Child Health, Sydney Medical School, The University of Sydney, Sydney, Australia.
7 Discipline of Paediatrics, School of Women’s and Children’s Health, UNSW Medicine, The University of New South Wales, Sydney, Australia.
2 Australian Paediatric Surveillance Unit, Kids Research Institute, Westmead NSW 2145, Australia.
Conflict of interest: None
3
Department of Respiratory Medicine, The Children’s Hospital at Westmead, Westmead NSW 2145, Australia. 4 Discipline of Paediatrics and Child Health, Sydney Medical School, The University of Sydney Clinical School, Sydney, Australia. 5
Correspondence to: Professor Adam Jaffe, School of Women’s and Children’s Health, Medicine UNSW, University of New South Wales, c/o Sydney Children’s Hospital, High Street, Randwick, New South Wales 2031, Australia E: [email protected] Received 30 June 2014; Revised 23 December 2014; Accepted 15 January 2015.
The Children’s Hospital at Westmead, Sydney, Australia.
6
Department of Respiratory Medicine, Sydney Children’s Hospital, Randwick NSW 2031, Australia.
ß 2015 Wiley Periodicals, Inc.
DOI 10.1002/ppul.23183 Published online in Wiley Online Library (wileyonlinelibrary.com).
2
Hime et al.
of chILD are breathlessness and hypoxemia in the presence of diffuse lung infiltrates on radiography or computed tomography. The term “diffuse lung disease” has often been used interchangeably, because the interstitial compartment is not always involved.1,2 Recently, the term “chILD syndrome” has been adopted by the American Thoracic Society Committee on chILD and the chILD Research Network in order to provide a set of criteria to aid the diagnosis of uncommon causes of diffuse lung disease phenotypes in infants.1 chILD differs from ILD in adults in that it is much rarer and, includes pathologies that are unique to infants and young children (e.g., pulmonary interstitial glycogenosis (PIG),6 neuroendocrine cell hyperplasia of infancy (NEHI)7, and chronic pneumonitis of infancy,8 and genetic abnormalities that cause ILD in childhood (e.g., surfactant protein deficiencies9). Conversely, common adult ILD such as usual interstitial pneumonia (UIP) is rarely described in children. Because individual disorders within the chILD group are extremely rare, this hinders research into this important area.10 Given the chronicity and high morbidity of chILD, the impact on families and burden on health services are likely to be significant. In this systematic literature review we aim to describe:
classification systems that have been applied to chILD the frequency of chILD in populations and the frequency of incident cases in hospitals
the morbidity, treatment, and outcomes of chILD the impact of chILD on families and the burden on health services
METHODS Identification of Relevant Studies
The following databases were searched: Medline (1946–2013), Embase (1966–2013), Evidence-Based Medicine (EBM) Reviews—Cochrane Database of Systematic Reviews (2005–2013), EBM Reviews— Cochrane Central Register of Controlled Trials (1991– 2013), Informit—Australasian Medical Index (1968– 2009), Informit—Health Collection (1977–2013), and clinicaltrials.gov (2000–2013). The following Medical Subject Headings (MeSH) were used in search strategies: lung diseases, interstitial; incidence; prevalence; public health; economics, medical; cost-benefit analysis; health services; health care rationing; health care costs. Forty text words (supplementary material, S1) were applied to searches independent of MeSH terms. The methodological filters “all child (0–18 years)” and “humans” were applied to all searches in Medline. NH and AP identified potentially relevant studies through screening of titles and abstracts. NH Pediatric Pulmonology
examined the full texts of potentially relevant studies and studies that fit the inclusion/exclusion criteria were included for review. To identify additional studies for inclusion, NH and AP screened the reference lists of review articles and relevant studies. Only articles in English were retrieved. Study Inclusion Criteria
Studies reporting original data in children aged 0–18 years, on the following aspects of chILD were included: classification systems, frequency (population prevalence/ incidence or hospital-based case series), symptoms, morbidities, treatments, outcomes, impacts on families, burdens on health services. Because chILD encompasses a heterogeneous group of disorders, this review is not limited to studies that included all chILD disorders but also includes studies that investigated specific chILD disorders. While we have used the currently accepted terminology of “chILD” in this systematic review, many of the epidemiological, and case series studies included for review were conducted prior to the development of the recent proposed classification schemes. Study Exclusion Criteria
Case studies reporting single patients or family groups were excluded. Studies involving adults (over 18 years of age) were excluded unless they reported sub-analyses of children. Studies reporting diagnostic evaluation without reporting on morbidity, treatment or outcome were excluded. Studies with insufficient detail to determine how and in whom the research was conducted were excluded. Also excluded from this review were common causes of diffuse lung disease that are not encompassed within “chILD syndrome”. As stated in the current clinical practice guidelines, these common causes include: cystic fibrosis, congenital, or acquired immunodeficiency, congenital heart disease, bronchopulmonary dysplasia, pulmonary infection, primary ciliary dyskinesia presenting with newborn respiratory distress, and recurrent aspiration.1 Data Extraction
NH and AP extracted data. Where possible, units of frequency were adjusted to facilitate comparisons between studies. RESULTS Literature search
A total of 37 studies of chILD disorders were included (Fig. 1).
4 presented classification systems for chILD 4 estimated population prevalence or incidence
Childhood Interstitial Lung Disease
3
included a group of disorders with onset of disease in infancy. Chronic aspiration syndromes, resolving acute respiratory distress syndrome and bronchopulmonary dysplasia were excluded. In the most recent chILD classification system, proposed by the American Thoracic Society Committee on chILD and the chILD Research Network, the term “chILD syndrome” was used to exclude common causes of diffuse lung disease such as cystic fibrosis, congenital heart disease, bronchopulmonary dysplasia, and pulmonary infection; it recognised that some chILD conditions may be asymptomatic when identified.1 It was acknowledged that many of the diffuse lung diseases (proposed by Deutsch et al.2) may present as chILD syndrome. Frequency of chILD in Populations and Hospitals
Fig. 1. Results of literature search strategy.
5 reported hospital-based incident cases of chILD within a specific time period
33 described morbidities, treatments, and/or outcomes Some studies reported data applicable to more than one of the categories above. We found no studies that reported impacts on families or burdens on health services. Classification Systems Applied to chILD
Four studies published between 2004 and 2013 describe classification systems for chILD (Table 1). There are inconsistencies in the definitions of chILD, each classification system included different disorders. Based on 185 cases of ILD in immunocompetent children in the 0–18 year age group, the European Respiratory Society Taskforce11 defined sub-groups according to aetiology and histopathology, modifying a classification of interstitial pneumonias in adults. Deutsch et al.2 classified 187 cases of diffuse lung disease using clinical data, imaging, and lung biopsy in children aged less than two years. They defined “diffuse lung disease” and included a broad range of diseases, some of which primarily affect distal airways outside of the interstitium. The classification had an emphasis on pathological findings and, disorders considered more prevalent in infancy were distinguished from other diffuse lung diseases. In 2010 Clement et al.5 proposed a step-wise approach to diagnosis of chILD in which specific disorders were grouped by aetiology. This classification
Three studies12–14 reported estimates of the frequency of chILD in populations (Table 2). Each study used different methods of case ascertainment and case definitions. Incidence estimates varied from 0.13 cases/ 100,000 children
