Abstract
Detailed literature searches were carried out in seven respiratory disease areas. Therapeutic evidence for efficacy of medicinal products was assessed using the Grades of Recommendation, Assessment and Evaluation (GRADE) methodology, as well as an assessment of safety and side-effects.
Systemic corticosteroids may reduce the development of bronchopulmonary dysplasia but have serious side-effects. Antioxidants need further study to demonstrate whether they have long-term benefits. Treatments for acute bronchiolitis have shown little benefit but new antiviral and monoclonal antibodies need further assessment. Well-constructed studies are needed to confirm the value of inhaled corticosteroids and/or montelukast in the management of viral-induced wheeze. Corticosteroids are the treatment of choice in croup. Minimal or no information is available for the treatment of congenital lung abnormalities, bronchiolitis obliterans and interstitial lung disease.
- Bronchiolitis
- bronchiolitis obliterans
- bronchopulmonary dysplasia
- croup
- interstitial lung disease
- viral-induced wheeze
There is a lack of evidence about the treatment of respiratory diseases that only occur in children. This report summarises work undertaken by a European Respiratory Society (ERS) task force, which successfully applied for funding to inform those caring for children with such diseases. The application was in parallel to a separate task force reviewing treatment of respiratory diseases occurring in children and adults. Seven categories of diseases were selected: bronchopulmonary dysplasia (BPD), sequelae of congenital lung abnormalities, acute viral bronchiolitis (AVB), croup, viral wheeze, bronchiolitis obliterans and paediatric interstitial lung disease (ILD), based on prevalence and priority for future research. The evidence for medicinal use in these seven diseases varies widely. European Union paediatric medicinal regulations have highlighted the lack of studies in children. The aims of this report are to summarise what evidence we have and to recommend where future research should be focused.
METHODS
Literature searches were run to identify material relating to the seven selected diseases. For each disease, a literature search was carried out based on a predefined series of keywords. Searching included the Cochrane library, MEDLINE and EMBASE, and the strategies included filters to limit the results by study type (reviews, randomised controlled trials (RCTs) and other types of experimental research) and by age range (0–18yrs). The details of the search strategies are available on request. In most cases the results were limited to English language material. No date limits were applied.
Each subgroup, consisting of two people, reviewed the retrieved references for each disease, adding additional papers from personal files if needed. The evidence from selected papers was graded, using the method described by the Grades of Recommendation, Assessment and Evaluation (GRADE) working group in 2004, as high-, moderate-, low- or very low-grade evidence, based on the following criteria: study design and quality (systematic reviews and RCTs: high quality; observational studies: low quality; any other type of article: very low quality), consistency of the data and relevance 1. A draft report was prepared by each subgroup which was submitted to the whole task force for comments. The reports were combined by the task force chairs (W. Lenney and F.M. de Benedictis) and recommendations were developed using the GRADE methodology. This enabled a more readily interpretable system of categorising recommendations in four groups: should (or should not) be done, or probably should (or should not) be done. Recommendations could only be categorised as “should (or should not) be done” when the entire task force unanimously endorsed this recommendation.
The task force found that the evidence on which to base recommendations is very limited for some diseases. When no evidence was available, narrative reviews and published expert opinions were considered for inclusion in the report.
BRONCHOPULMONARY DYSPLASIA
Definition
Pre-term babies still requiring additional oxygen at gestational age 36 weeks.
Prevention of BPD: the evidence base
Surfactant
Since treatment with exogenous surfactant was first reported 2, studies have shown it reduces mortality and air leaks 3–12. In one trial, synthetic surfactant was associated with a reduction in BPD development 13. Meta-analysis of five RCTs also demonstrated synthetic surfactant in established respiratory distress syndrome (RDS) reduced the risk of BPD 3.
Vitamin A
Vitamin A is needed in the fetal lung for cellular differentiation and surfactant synthesis. Very low birth weight infants are frequently deficient in vitamin A. Several RCTs have been undertaken to determine its efficacy to prevent BPD. A Cochrane review reported vitamin A supplementation reduces death or oxygen requirement at 1 month of age and oxygen requirement at 36 weeks 14. In six of the studies in the review, vitamin A was given by intramuscular injection 15–20, in two it was given orally 21, 22. The dose and the frequency varied 23. High levels of vitamin A can cause neurological signs and symptoms. Neuro-developmental outcome is only available for one study 23. The optimum vitamin A dose and route of administration require further studies.
Systemic corticosteroids
Dexamethasone
The efficacy of early administration (within 96 h of birth) has been examined in 21 studies showing a reduction in BPD development 24. Seven studies examined the results of administration between 7 and 14 days showing a reduction in BPD and ability to extubate 25. Meta-analysis of 20 RCTs studying 2,064 infants showed the relative risk for cerebral palsy was increased in those receiving early treatment 26; therefore, use of systemic corticosteroids cannot be recommended.
Various strategies have tried to improve the risk/benefit ratio, including a lower dosage. Infants of <28 weeks’ gestation were randomised to receive 10 days of dexamethasone 0.89 mg·kg−1 or placebo 27. BPD development was unchanged but dexamethasone allowed earlier extubation. No difference in death or major disability was seen at 2 yrs 28.
Hydrocortisone
A pilot study suggested early low-dose hydrocortisone might increase survival without BPD 29. A multicentre RCT, however, was halted prematurely due to an increased rate of gastrointestinal perforation in the hydrocortisone-treated infants. Overall, there was no significant difference in BPD-free survival and mortality but hydrocortisone was associated with a lower mortality and improved survival without BPD in the subgroup exposed to chorioamnionitis 30. In another RCT, oxygen-free survival was higher in infants who received hydrocortisone 31. The latter study, however, was halted prematurely and no conclusion can be drawn regarding the efficacy and safety of hydrocortisone.
Inhaled corticosteroids
Meta-analysis of the results of 11 RCTs in which inhaled corticosteroids were given before 2 weeks of age showed no reduction in BPD development 32.
Nitric oxide
Inhaled nitric oxide (iNO) selectively decreases pulmonary vascular resistance and improves oxygenation. It reduces the need for extracorporeal membrane oxygenation in newborns with severe hypoxic respiratory failure and pulmonary hypertension 33.
It was proposed that iNO would decrease the need for a high oxygen requirement and ventilatory support in pre-term infants with RDS. It can prolong bleeding time 34 and inhibit platelet aggregation 35, hence the concerns regarding intracranial haemorrhage 36.
A systematic review has been performed of 11 RCTs which used iNO 37. The studies were separated into three groups. Meta-analysis of seven RCTs demonstrated that iNO given as early rescue therapy in the first 3 days of life resulted in no significant effects on BPD development or mortality, but there was a trend towards an increased risk of severe intracranial haemorrhage. Later use of iNO resulted in no significant effect on BPD. In one trial, there was a reduction in death or BPD in infants aged 7–14 days at randomisation 38. Routine use of iNO in intubated prematurely born infants resulted in a marginal significant reduction in the combined outcome of death or BPD and a reduction in severe intracranial haemorrhage or periventricular leukomalacia. There is no evidence from RCTs that iNO has a detrimental effect on neurodevelopmental outcome in prematurely born infants 39, 40.
Nitric oxide has anti-inflammatory effects and promotes cell and vessel growth in the immature lung 41. It may improve surfactant function without altering surfactant recovery or protein composition 42. Therefore, earlier iNO therapy, before lung disease is established, might prevent BPD in less sick prematurely born infants 37, 43, 44. The results of a large pharmaceutical (INO Therapeutics, Clinton, NJ, USA)-sponsored trial of prophylactic low-dose iNO are awaited.
Other possible preventative therapies
Inositol
Inositol promotes maturation of surfactant phospholipids and the synthesis of phosphatidyl inositol in type 2 pneumocytes. Meta-analysis of two RCTs demonstrated that inositol supplementation significantly reduced the combined outcome of death and BPD, but not BPD alone 45.
Bronchodilators
There has only been one RCT and this showed no benefit 46.
Cromolyn sodium
No reduction in BPD was demonstrated in two small RCTs 47.
Macrolide antibiotics
Erythromycin. The relative risk of BPD development in babies colonised with Ureaplasma urealyticum is 1.7 48. Administration of erythromycin did not reduce this 49 but duration of therapy may have been too short 50.
Azithromycin. Macrolide antibiotics have anti-inflammatory actions and inhibit neutrophil chemotaxis and superoxide generation. In an animal model of hyperoxia, treatment with azithromycin improved survival and decreased lung damage 51. It also reduced interleukin (IL)-6 and IL-8 production by tracheal cells obtained from prematurely born infants 52. Azithromycin, however, did not reduce the incidence of BPD in infants <1,001 g but was associated with less postnatal steroid use 53.
Antioxidants
Superoxide dismutase. Lung injury secondary to hyperoxia can be reduced by superoxide dismutase (SOD) 54, 55. Prophylactic supplementation with recombinant human copper–zinc SOD reduced repeated severe wheeze during infancy, but did not reduce BPD development 56.
N-acetyl cysteine. The synthesis of glutathionine (an endogenous scavenger of free radicals) is limited by the availability of cysteine. In a multicentre RCT, no significant differences were found in the incidence or severity of BPD 57 or in lung function 58. N-acetyl cysteine, however, was started at age 36 h 57, which may have been too late.
Allopurinol. Xanthine oxidase generates superoxide radicals following hypoxia. Allopurinol is a synthetic inhibitor of xanthine oxidase and a free radical scavenger. In an RCT of 400 extremely premature infants, allopurinol did not reduce BPD incidence 59.
Melatonin. Melatonin, a potent free radical scavenger, upregulates antioxidant enzymes and downregulates pro-oxidant enzymes 60. Infants with RDS treated with melatonin had lower levels of pro-inflammatory cytokines and reduced ventilatory requirements 61.
Vitamin E. Vitamin E is a scavenger of free radicals which inhibits inflammation. Meta-analysis demonstrated that vitamin E supplementation did not reduce the risk of BPD; a systematic review of 26 RCTs showed high-dose intravenous vitamin E supplementation increased the risk of sepsis 62.
Cimetidine. Cimetidine, a cytochrome P450 inhibitor, prevents gas exchange failure in newborn lambs after breathing 95% oxygen for 72 h 63. In an RCT of 84 newborn infants <1,251 g, cimetidine had no effect on severity of respiratory insufficiency at 10 days 64.
Methylxanthines
Caffeine. Caffeine administration has been associated with a reduction in BPD development 65. Follow-up demonstrated a reduction in cerebral palsy 66.
Pentoxiphylline. Pentoxiphylline, a methylxanthine derivative and nonselective phosphodiesterase inhibitor, has anti-inflammatory effects. It also inhibits synthesis of intercellular adhesion molecule (ICAM)-1, which correlates with the risk of developing BPD 67. In rat pups it reduced fibrin deposition and prolonged survival in experimental hyperoxic lung injury 68. In five infants with BPD, nebulised pentoxiphylline reduced oxygen requirements within 5 days 69. In an RCT of 150 very low birth weight infants, pentoxiphylline was associated with a 27% reduction in BPD development 70.
Oestradiol and progesterone
Oestradiol and progesterone are important in lung growth 71, 72. In infants of <29 weeks’ gestation, they were given for ≥2 weeks; 48% of the placebo and 44% of the hormone group developed BPD or died, but the longer the hormones were given, the lower was the risk of BPD 73.
Ethyl nitrate
Inhaled ethyl nitrate achieves S-nitrosation of glutathione without releasing reactive nitrogen species 74. It had a greater effect than nitric oxide in preventing lung myeloperoxidase accumulation and expression of cytokine-induced neutrophil chemoattractant in newborn rats 75. Human studies in BPD are under consideration.
Prevention of BPD: summary and recommendations
For a summary of recommendations for the prevention of BPD, see table 1⇓.
Prevention of bronchopulmonary dysplasia(BPD)
Side-effects
1) Early (within 96 h of birth) systemic corticosteroids increase the risk of cerebral palsy, so routine use in the first weeks of life cannot be recommended.
2) High levels of vitamin A can cause neurological signs due to raised intracranial pressure.
Conclusions
1) Many therapies have been studied in RCTs but no safe and effective BPD preventative therapy has been shown.
2) Systemic corticosteroids, in the first weeks after birth, reduce BPD, but have serious side-effects.
3) Vitamin A reduces BPD but also has side-effects. Further investigation is needed to identify the safest dose regimen.
Prioritised research questions
1) Antioxidants, such as SOD, and early low-dose iNO require further studies to determine if they can prevent BPD.
2) The positive effect of caffeine on BPD incidence is exciting. Optimum timing and dosage related to long-term outcomes merit testing.
Treatment of BPD: the evidence base
Surfactant
Ventilated pre-term infants have abnormal surfactant function with deficiency of surfactant protein A and B 76. They also have abnormalities of surfactant kinetics, which correlate with the level of ventilatory support 77.
There have been very few studies examining surfactant use in infants with BPD. A single dose was given to 10 ventilated infants of 25 weeks’ gestation. Oxygenation improved for 24 h 78. Six infants who had received a repeat surfactant course were retrospectively identified; five subsequently developed BPD 79.
Systemic corticosteroids
Infants with BPD have persisting lung inflammation. In the Cochrane review assessing nine RCTs (562 infants) in which corticosteroids were given, no effect on mortality was shown 80. One study showed a reduction in oxygen dependence at 36 weeks associated with less need for home oxygen. Adverse effects included glycosuria and hypertension, and slight increase in retinopathy of prematurity. No study has been powered to detect adverse long-term neurosensory outcomes. It is prudent to reserve use of late corticosteroids for infants who cannot be weaned from mechanical ventilation.
Diuretics
The efficacy of diuretics, including aerosolised diuretics, in infants with developing or established BPD has been assessed 81–94. RCTs have shown short-term benefits in lung mechanics and/or oxygenation, but there have been no significant effects on long-term outcomes.
Bronchodilators
Inhaled bronchodilator therapy improves lung function and blood gases in ventilated 1-month-old babies with BPD 95. Salbutamol resulted in a dose-dependent improvement in lung mechanics 96. Synergism was seen with salbutamol and ipratropium bromide 97, but not metaproterenol and atropine 98. Any effect was short-lasting.
Pulmonary vasodilators
Pulmonary hypertension is associated with BPD and is a recognised cause of mortality 99–101. Abnormally high vascular reactivity is present in BPD with a marked increased vasoconstrictor response to acute hypoxia 102.
Calcium channel blockers, prostacyclin, sidenafil and endothelin-1 antagonists are potent pulmonary vasodilators and are used to treat pulmonary hypertension in infants with BPD. No RCTs have been undertaken, with most studies comprising few or single patient case reports 103–108.
Nitric oxide
iNO was given for 72 h to ventilator-dependent BPD infants >4 weeks old. 11 responded, and four of the 11 were ultimately weaned off mechanical ventilation. All infants who failed to respond either died or continued to require mechanical ventilation 109.
Treatment of BPD: summary and recommendations
For a summary of recommendations for the treatment of BPD, see table 2⇓.
Treatment of bronchopulmonary dysplasia(BPD)
Side-effects
1) Corticosteroids increase glycosuria and hypertension; no study has been powered to detect adverse long-term neurodevelopmental outcomes.
2) Chronic diuretic therapy may cause hypercalcuria, renal calcification and nephrolithiasis leading to haematuria and urinary tract infections.
Conclusions
1) Corticosteroids should only be given to infants with severe respiratory failure and only continued beyond 3 days if there is obvious clinical response.
2) Diuretics should be reserved for infants with fluid overload.
3) Bronchodilators should be given to infants who have symptomatic wheeze.
Prioritised research questions
1) The efficacy of various pulmonary vasodilators in improving long-term outcome in BPD needs investigation.
2) The impact of BPD treatments on long-term lung function should be an essential outcome of all RCTs.
CONGENITAL LUNG MALFORMATIONS
The possibility of medical treatment: the evidence base
Although uncommon individually, congenital lung cysts, congenital cystic adenomatous malformations, lung sequestrations, congenital lobar emphysema, and tracheal and airway narrowings collectively pose management difficulties in early childhood. Post-natal therapy is either surgical excision, coil embolisation of feeding vessels, or expectant treatment, with no definite evidence to guide the paediatrician. There are no known medical therapies which lead to regression of congenital malformations. Antenatal treatment with betamethasone of mothers carrying a fetus with a congenital thoracic malformation has led to regression of the malformation and better results than historical controls 110, 111. However, an unexpected intrauterine death associated with this treatment 112 means it is unlikely to become routine. There are no other antenatal medical therapies which are likely to be testable in the near future.
Prioritised research questions
There are none relating to medications at present.
ACUTE VIRAL BRONCHIOLITIS
Treatment of AVB: the evidence base
AVB is frequently caused by the respiratory syncytial virus (RSV) 113. Hospitalisation rates are ∼1%, but can be 10–20% in high-risk groups 114–116. About 10% hospitalised infants require mechanical ventilation 117. Mortality is very low. AVB is followed by recurrent episodes of wheeze in 30–70% of young children.
Bronchodilators
A meta-analysis concluded that bronchodilators have no effect on the course of AVB 118. A large RCT 119 and a Cochrane review 120 stated that there is insufficient evidence to support the use of epinephrine in inpatients; however, epinephrine may be helpful in the outpatient setting 120. Individual trial of nebulised bronchodilators may be justified, but the medication should be discontinued unless a clear positive effect is seen 121–123.
Inhaled corticosteroids
No effect of inhaled corticosteroids was seen on clinical scores or length of hospitalisation 124–126. There is no evidence that inhaled corticosteroids prevent recurrent wheeze following AVB 127, 128. There are no large RCTs focusing on the long-term effects of inhaled corticosteroids in children with AVB.
Systemic steroids
A meta-analysis concluded that there is no positive effect of systemic corticosteroids on the course of AVB 129. An RCT in hospitalised infants showed modest benefit of a single dose of dexamethasone 0.6 mg·kg−1 on several outcomes, including length of hospital stay 130. A large RCT did not show any effect of one dose of dexamethasone in 600 infants presenting at the emergency department 131. One study showed benefit of i.v. dexamethasone in children mechanically ventilated for life-threatening AVB 132.
Leukotriene receptor antagonists
Increased concentrations of leukotrienes have been found in the airways of infants with AVB 133. A recent study showed that montelukast does not improve the clinical course of AVB 134. A preliminary study showed a beneficial effect of montelukast on post-bronchiolitis wheeze 135, but this was not confirmed in a larger study 136.
Polyclonal and monoclonal antibodies
Antibody treatment might prevent bronchiolitis in high-risk populations. Polyclonal immunoglobulins against RSV are no longer available. A Cochrane review did not show benefit in four RCTs of polyclonal immunoglobulins used as AVB treatment 137.
Palivizumab is effective as RSV prophylaxis and is used in many countries in patients at high risk of developing severe RSV disease 138. The recommended regime is five i.m. injections at monthly intervals over the RSV season. The major restriction to its use is its high cost.
Antibiotics
Bacterial infection is rare during AVB 139. In 1966, Field et al. 140 reported no positive effect of broad-spectrum β-lactam antibiotics on the course of disease in 52 patients. In 2007, Tahan et al. 141 reported that macrolides reduced length of hospitalisation. A Dutch RCT did not show that azithromycin reduces the duration of hospitalisation 142. The anti-inflammatory effect of macrolides should be studied in RCTs in outpatient and inpatient settings.
Antiviral drugs
Small studies have been performed with ribavirin 143, 144. The outcomes are variable on length of hospital stay, duration of mechanical ventilation and effect on post-bronchiolitic wheeze. In the absence of large RCTs the effect of ribavirin remains unproven.
Chest physiotherapy
A systematic review of physiotherapy in children aged <24 months has been performed 145. The studies used different vibration and percussion techniques. There was insufficient evidence to undertake a meta-analysis.
Hypertonic saline
Three small studies evaluated the effect of aerosolised hypertonic saline (3%) in infants 146–148. In ambulatory patients it reduced symptom score 146. In hospitalised infants it reduced the length of hospital stay 147. A similar reduction of hospital stay was reported in infants receiving 3% saline solution versus those receiving normal saline in addition to routine therapy 148. The beneficial effect, the lack of side-effects and the limited cost of this treatment deserve consideration for a large RCT.
Treatment of AVB: summary and recommendations
Recommendations for treatment of AVB can be found in table 3⇓.
Treatment of acute viral bronchiolitis
Side-effects
The interventions mentioned in this article have no important side-effects.
Conclusions
AVB is a frequent cause of morbidity during early childhood. Supportive care, including administration of oxygen and fluids, is the cornerstone of treatment. There is no specific treatment with clear beneficial effect on the course of disease. Differences in clinical definition of AVB exist between countries 121, 149. Uniformity is needed to better differentiate infants with AVB from those with other illness, and to allow a comparison between studies performed in different parts of the world. Inpatient and outpatient studies should be considered separately. In inpatient studies, length of hospital stay is important. In outpatient studies, prevention of admission should be the primary outcome.
Prioritised research questions
1) What are the safety and efficacy benefits of antiviral drugs and monoclonal antibodies in the treatment of AVB?
2) Can the beneficial effect of hypertonic saline be confirmed in large RCTs?
3) What is the effect of combined antiviral and immunomodulatory treatments on short- and long-term outcomes in AVB?
VIRAL-INDUCED WHEEZE
Treatment of viral-induced wheeze: the evidence base
The term “viral-induced wheeze” is a much used entity in pre-school children defined as “wheezing only in association with viral respiratory tract infection and with no interval symptoms” 150. However, there may be difficulties in discriminating the atopic asthmatic pre-school child from this definition 151.
Wheeze is a word found only in the English language. It is often described as a whistling noise in the airways. In this section the terms with respect to treatment will be defined as follows. “Acute viral-wheeze”: an episode or attack of wheeze in association with a clinical viral respiratory tract infection, not dependent on previous history wheeze; and “episodic viral wheeze”: wheezing only in association with viral respiratory tract infection with no interval symptoms but dependent on clinical history.
Spycher et al. 152 published an assessment of pre-school wheeze in children from a population-based cohort. They identified five phenotypes, one being transient viral wheeze. A problem with such retrospective classification is that it is not applicable in the acute situation when the child has symptoms.
There are issues with the definition of acute viral wheeze in infants. In the USA and some European countries, infants with a major wheeze component triggered by a viral cold may be classified as having “bronchiolitis” 153. In the UK, the diagnosis “bronchiolitis” is usually reserved for infants with the combination of acute bilateral chest crackles, wet cough, hypoxaemia and poor feeding 154. In this review, it is assumed that some of the non-UK studies have included infants with a mix of acute viral wheeze and UK-defined “bronchiolitis”.
Respiratory viral infections and bronchial obstruction/wheeze
In the early 1970s, RSV was established as causing winter epidemics of acute bronchiolitis 155, 156. The different clinical spectrum of RSV was recognised 157. Other respiratory viruses, such as parainfluenza, rhinoviruses, influenza and adenovirus were recognised as other causes 158. Later, human metapneumovirus was identified as another cause 159. The relationship of rhinovirus infections to acute exacerbations of asthma was also noted in older children 160. ICAM-1 is the major receptor for human rhinoviruses 161. RV infections early in life might predispose to later asthma development 162, but a more probable explanation is that this virus infection may affect persons predisposed to have asthma later in life.
This section offers an overview of the treatment of viral-induced wheeze, both as treatment of a single episode and of recurrent episodes. It has been assumed the majority of episodes will be triggered by a viral cold. Realistically, clinicians treating an acute wheeze will not stratify treatment based on a parent who may be unable to give an objective history of the long-term pattern of wheeze. For outpatient therapy, clinicians and parents have the time to assess that pattern of wheeze and to assign a putative phenotype. For regular therapy, this section separates studies of children with intermittent wheeze from those with interval symptoms between attacks. Such separation has been clearly incorporated into the recent ERS task force on wheezing disorders in pre-school children 163.
Treatment of acute viral wheeze
Stratification of treatment based on wheeze phenotype (i.e. episodic viral wheeze versus multi-trigger wheeze) is assumed to be difficult in the emergency room.
Short-acting β2-agonists
Short-acting β2-agonists are the main treatment for wheeze, both parent-initiated (i.e. in the community) and physician-initiated. The evidence for efficacy falls broadly into two categories. First, lung function changes are “objective”, but clinical relevance may be unclear. Secondly, there are studies using clinically relevant outcomes in acutely wheezy children. A 2002 Cochrane review sought to determine the effectiveness of β2-agonists for infants (<2 yrs) with “recurrent or persistent wheeze” 164. Eight studies meeting the Cochrane criteria were identified 165–172. Five studies had clinical end-points, and three assessed changes in lung function. Overall, no clear benefit for the use of β2-agonists was found. The inconsistent benefit of short-acting β2-agonists has also been seen in studies since 2002. Skoner et al. 173 found no benefit of parent-given nebulised albuterol/levalbuterol in children aged 2–5 yrs with mild wheeze, in whom the primary outcome was change in caregiver asthma score. In a laboratory setting, however, lung function (specific airway resistance) improved significantly after inhaled salbutamol 174.
The evidence from well-conducted studies that inhaled short-acting β2-agonists result in clinically relevant outcomes (such as wheeze severity score) in young children (<2 yrs) with acute viral wheeze remains weak. Since some beneficial effects of β2-agonists have been reported for lung function, it is very unlikely that an RCT of inhaled short-acting β2-agonist therapy in older children with clinically severe wheeze (for either phenotype) will be ethically acceptable, and this approach will remain the cornerstone of acute treatment.
One factor that may underlie the inconsistent results of short-acting β2-agonist studies in the pre-school age group is a combination of reduced dose delivered to the lower airway, and/or reduced responsiveness to β2-agonists. Intravenous salbutamol is one way of ensuring that a high concentration of this drug is delivered to the lower airway, albeit with the possibility of increased systemic side-effects. Browne et al. 175 assessed 15 μg·kg−1 i.v. salbutamol over 10 min in acutely wheezy children who did not improve after a single dose of nebulised salbutamol: i.v. treated children were ready for discharge 9.7 h earlier than controls, with no additional side-effects; responses in the pre-school subgroup were not assessed separately, the majority of children being of school age. Assessing the efficacy of i.v. salbutamol as an adjunct to inhaled β2-agonist therapy in acutely wheezy pre-school children is ethically acceptable but no trial has addressed this.
Inhaled epinephrine
Inhaled racemic epinephrine is used for wheeze associated with AVB. A large RCT and a Cochrane review stated that there is insufficient evidence to support the use of epinephrine in the treatment of inpatients with bronchiolitis, but there is some evidence that epinephrine may be favourable to salbutamol and placebo among outpatients 119, 120.
In summary, inhaled epinephrine may have a short-term effect upon symptoms of bronchial obstruction in studies that probably have recruited infants with a mix of “bronchiolitis” (UK definition) and acute viral wheeze (using our criteria). Further studies of nebulised adrenaline should aim to recruit pre-school children aged 1–5 yrs, in whom RSV bronchiolitis (UK definition) can be excluded because of the older age of the children and, perhaps, the lack of widespread chest crackles.
Intermittent high-dose inhaled corticosteroids
In a systematic review 176, three high-quality studies of intermittent high-dose inhaled corticosteroids for the treatment of the episodic viral wheeze phenotype were identified 177–179. Steroids were started at the onset of the acute episode. Only a modest improvement in symptoms was achieved. Ducharme et al. 180 found a modest benefit of high-dose fluticasone (3 mg beclomethasone dipropionate equivalent) in a 6–12 month RCT of 129 pre-school children with a history of episodic viral wheeze. Fluticasone or placebo was started at the first sign of a cold and continued for up to 10 days. In the active treatment group, 8% of colds resulted in the need for rescue oral steroids compared with 18% in the placebo group. Pre-emptive steroid treatment was associated with smaller mean increases in height and weight. No study has assessed the role of high-dose intermittent inhaled corticosteroids in children with interval symptoms and viral-triggered attacks.
Oral corticosteroids
A trial of parent-initiated oral steroids therapy (PIOST) for intermittent pre-school wheeze, in children who had previously been hospitalised with an attack of acute viral wheeze, found no difference in symptom scores and the number of salbutamol actuations per day 181. Issuing parents with a course of oral corticosteroids to be given at the first sign of an attack of viral wheeze does not, therefore, appear to be an effective strategy. In contrast, in children aged 6–35 months with acute viral wheeze, oral prednisolone 2 mg·kg−1·day−1 for 3 days given at presentation to the paediatric emergency department, reduced disease severity, length of hospital stay and the duration of symptoms in those who were subsequently hospitalised 182. An RCT of 700 children presenting to hospital with acute viral wheeze (most with a history of episodic viral wheeze) found that a short course of prednisolone did not reduce length of stay in hospital or reduce objective measure of wheeze severity 183. Meta-analysis of data from these studies is, therefore, urgently required in order to answer the question of steroid efficacy in viral wheeze. However, the results of the latest trial suggest there may be little or no benefit for the majority of children with acute viral wheeze. Future studies should assess whether there is a steroid-responsive subgroup.
Leukotriene receptor antagonists
In a recent study in pre-school children with moderate-to-severe intermittent wheezing, episodic use of either montelukast or budesonide early in respiratory tract illness, when added to albuterol, did not increase the proportion of episode-free days or decrease oral corticosteroid use over a 12-month period. However, indicators of severity of acute illnesses were reduced, particularly in children with positive asthma predictive indices 184.
Regular “prophylactic” treatment
Inhaled corticosteroids
A systematic review 176 identified only one RCT of regular inhaled corticosteroids for the episodic viral wheeze phenotype in pre-school children. In this study, Wilson et al. 185 compared the effect of 4 months’ regular treatment with budesonide 400 μg·day−1 with placebo and found no difference in symptom score, oral steroid usage or in hospital admission between the two groups. Kaditis et al. 186 identified 10 placebo-controlled trials in pre-school children with “persistent” wheezing (377 subjects in total) who received placebo or regular inhaled corticosteroids. Inhaled corticosteroids decreased the symptom score, β2-agonist use and oral steroid requirement and increased the mean peak expiratory flow rate. Hospitalisation was not reduced. Kaditis et al. 186 concluded that the effects may be clinically trivial. There are major problems with assessing trials that have recruited a “mix” of phenotypes. A Cochrane approach performed by researchers with a detailed knowledge of pre-school wheeze phenotypes is the only way of avoiding bias, but this has not yet been done.
Long-acting β2-agonists
Regular inhaled long-acting β2-agonists (LABAs) in school-aged children are effective when given with regular inhaled corticosteroids. Bronchodilation with salmeterol is not immediate. Formoterol has a more rapid onset. There are no clinical data for either drug in studies that have recruited only pre-school children. In a Cochrane review, Ni Chroinin et al. 187 assessed LABA versus placebo in addition to inhaled corticosteroids in “children and adults with chronic asthma”. Eight of the 26 included trials focused on children. The authors concluded “there are insufficient data to make firm conclusions on the use of LABA for pre-school and school-aged children.” Evidence that formoterol may achieve prolonged bronchodilation in pre-school children is provided by Nielsen and Bisgaard 174 who found that it produced sustained bronchodilation for at least 8 h in pre-school children who were not wheezing at the time of study.
Respiratory function studies support the use of LABA in some pre-school children, but no trial has assessed its use as an “add-on” therapy to inhaled corticosteroids in pre-school asthma using clinically relevant outcomes. By definition, no study has assessed the efficacy of add-on LABA in preventing viral-triggered wheeze attacks in the intermittent wheeze phenotype.
Leukotriene receptor antagonists
Bisgaard et al. 188 investigated the effect of continuous montelukast therapy on asthma exacerbations in children aged 2–5 yrs, with a history of “intermittent wheezing” and minimal or no interval symptoms. Children were randomised to receive either 4 mg oral montelukast or placebo once a day for 12 months. Montelukast reduced the rate of asthma exacerbations, inhaled corticosteroid and β-agonist usage, but authors admitted that mild, intermittent symptoms should not be treated daily with montelukast and management should not necessarily involve regular treatment all year round. A study from Australia in children aged 2–14 yrs has suggested that a short course of montelukast, introduced at the first signs of a wheezing episode, results in a modest reduction in healthcare resource utilisation, symptoms, time off from school and parental time off from work in children with intermittent symptoms 189.
In summary, there is some evidence of efficacy of montelukast as monotherapy in episodic viral-wheeze phenotype and possibly for “multi-trigger persistent” pre-school wheeze. Its role as an add-on therapy for pre-school children receiving inhaled corticosteroids is less clear. A significant level of nonresponsiveness has been shown in older children with regular asthma 190, 191, but the frequency of nonresponse has not been reported in young children with wheeze. One should be aware of the recent large randomised placebo-controlled study of montelukast treatment for post-RSV bronchiolitic wheeze (in 979 patients), which showed no effect 192.
Treatment of viral-induced wheeze: summary and recommendations
Recommendations for treatment of viral-induced wheeze can be found in tables 4⇓ and 5⇓.
Treatment of acute pre-school viral wheeze
Regular preventative therapy for episodic pre-school viral wheeze
Side-effects
Medicines used to treat viral induced wheeze are considered safe but some aspects should be kept in mind. The Food and Drug Administration (USA) has warned against the use of inhaled LABA without concomitant anti-inflammatory treatment. Oral corticosteroids have systemic side-effects when used for long or repeated periods of time. Also, high-dose inhaled corticosteroids may affect growth and the hypothalamic–pituitary–adrenal axis.
Conclusions
The evidence base for treatment of acute viral wheeze and for regular episodic viral wheeze phenotype is weak. Inhaled corticosteroids decrease symptom score, β2-agonist requirement and oral corticosteroid use, and increase mean peak flow rates in some but not all studies. One problem is that the phenotype is often not clearly described.
Prioritised research questions
1) Consensus on the definition of pre-school wheeze phenotypes is needed as the basis for clinical trial protocols. Outcomes must be clearly defined.
2) What are the benefits of high-dose inhaled corticosteroids, oral montelukast and i.v. salbutamol for severe acute viral-induced wheeze?
3) What are the benefits of intermittent oral theophylline or intermittent oral montelukast in recurrent wheeze in pre-school children?
CROUP
Treatment of croup: the evidence base
Viral croup is the most frequent cause of acute upper airway obstruction in children 6 months to 6 yrs of age; parainfluenza viruses types 1 and 3 account for the majority of cases. Croup symptoms are generally short-lived; the majority of children show resolution within 2 days. Most have a mild illness, but pronounced laryngeal obstruction can lead to asynchronous chest wall and abdominal movements, fatigue, hypoxia, hypercapnia and respiratory failure 193. The mortality rate is very low 194. Treatment has evolved from mist kettles and tents to the evidence-based practice of corticosteroids and nebulised epinephrine. Very recently, two reviews on the topic were published 195, 196.
Humidified air
There is a long history of using humidified air (mist therapy) in the treatment of croup. A systematic review of three RCTs concluded that, in children with mild-to-moderate croup, there is no evidence that inhalation of humidified air improves the croup score 197. In another RCT, the effects of 100% humidified oxygen, 40% humidified oxygen and standard oxygen (humidity as in ambient room air) were compared in children with moderate-to-severe croup 198. There were no differences in croup score, treatment with dexamethasone or epinephrine or admission to hospital between the three groups.
Corticosteroids
There is clear evidence for the effectiveness of steroids, which are now routinely recommended. Meta-analyses of RCTs 199–201 have consistently demonstrated significant improvements in patients treated with corticosteroids compared with controls. Corticosteroids decrease the time spent in the emergency room or hospital, the number of return visits and hospital admissions. They also reduce the use of epinephrine, the need for intubation and the risk of reintubation 199–202.
The routes of steroid administration have been extensively investigated. The oral or i.m. routes have been shown to be either equivalent or superior to inhalation 203–206. The addition of inhaled budesonide to oral dexamethasone was shown to offer no advantage in hospitalised children 207.
Two trials comparing oral and i.m. dexamethasone noted no difference with regard to resolution of croup symptoms, admission to hospital or further treatment 208, 209. A study comparing i.m. dexamethasone to oral betamethasone also found no difference in reduction of croup score, hospital admission, time to symptom resolution or return for medical care 210. Two studies compared oral dexamethasone to oral prednisolone: dexamethasone was found to be superior to prednisolone in reducing rates of return for medical care in one 211; no difference was noted in reduction of croup score or rates of return for medical care in the other 212. No studies have directly compared the outcomes of single- and multiple-dose therapy.
There is conflicting evidence with regard to the effect of dose size of steroids. In a meta-analysis 199, more children admitted to hospital responded to treatment when higher doses of hydrocortisone equivalents were used. A few small RCTs compared different doses of oral dexamethasone and suggested that a dose of 0.15 mg·kg−1 was as efficacious as 0.3 or 0.6 mg·kg−1 212–214.
The mechanism of action of steroids in croup is not well known. The rapidity of action suggests steroids probably act through vasoconstriction in airway mucosa 215.
Epinephrine
Nebulised racemic epinephrine has been studied over three decades. Nebulised epinephrine by intermittent positive pressure breathing improves croup scores within 30 min of treatment but the benefit lasts for <2 h 216, 217. Epinephrine by nebulisation alone was found to be as effective as nebulisation using intermittent positive pressure breathing 218. Substantial improvement after treatment with epinephrine has also been shown in five prospective studies investigating various objective measures of severity; the clinical effect was found to be sustained for at ≥1 h 219–224. An RCT showed l-epinephrine 1:1,000 to be as effective as racemic epinephrine 225.
The effect of epinephrine has been compared with nebulised budesonide in 66 hospitalised children with moderately severe croup; there was no significant difference in efficacy and safety for several outcomes 226.
Heliox
Theoretically, the administration of a helium–oxygen mixture (heliox) could reduce the degree of respiratory distress as helium, with a lower density than nitrogen, can pass through a narrowed airway with less turbulence. Two RCTs, one comparing heliox to epinephrine in children who had already received i.m. dexamethasone 227, and another comparing heliox to oxygen-enriched air in children with mild croup 228, did not show significant differences in croup scores between the groups. Due to costs and practical limitations, the general use of heliox cannot be recommended.
Other treatments
Children with croup should be made as comfortable as possible because agitation may cause substantial worsening of symptoms. Children with moderate or severe croup and hypoxia should receive oxygen. Analgesics or antipyretics may be used to reduce fever or discomfort. The use of antitussives, decongestants, antibiotics and β2-agonists is not indicated.
Treatment of croup: summary and recommendations
Recommendations for treatment of croup can be found in table 6⇓.
Treatment of acute viral croup
Side-effects
Corticosteroids are considered safe in short courses. Concerns exist with respect to their immunosuppressive effects, which might predispose to infectious complications 195, 196. Nebulised epinephrine has not been associated with major adverse effects 196.
Conclusions
Corticosteroids are the treatment of choice for children with croup. Benefit is seen at all levels of severity. Nebulised epinephrine is effective for temporary relief of symptoms of airway obstruction and may be used until steroids take effect. Mist therapy has been shown to be ineffective.
Prioritised research questions
1) To compare the efficacy of a single dose of oral dexamethasone with a single dose of oral prednisolone in children with mild-to-moderate croup.
2) To compare the outcomes of single- versus multiple-dose therapy.
POST-INFECTIOUS BRONCHIOLITIS OBLITERANS
Treatment of post-infectious bronchiolitis obliterans: the evidence base
Bronchiolitis obliterans refers both to histological findings and to a clinical syndrome of airflow obstruction. It is the end stage of a process that begins with bronchiolar epithelial injury followed by an inflammatory reaction which leads to airway obliteration.
Bronchiolitis obliterans is defined histologically by the presence of granulation tissue within small airways and/or destruction of the small airways by fibrous tissue. The typical pathological changes have a patchy distribution 229.
Respiratory infection is the commonest cause of bronchiolitis obliterans in children (post-infectious bronchiolitis obliterans (PIBO)). Other causes include inhalation of noxious gases, drug toxicity and connective tissue disease 230–234. It is also seen following lung transplantation where it is referred to as bronchiolitis obliterans syndrome. Infection with adenovirus (particularly serotypes 3, 7 and 21) is by far the most common cause of PIBO. Other infectious agents include respiratory syncytial virus, influenza virus, parainfluenza virus, measles virus, Mycoplasma pneumoniae, Chlamydophylia pneumoniae and Bordatella pertussis 230–233.
The prevalence of PIBO is not known and varies in different parts of the world. In Western Europe it is relatively rare, with most centres only expecting to see one or two cases per year. The prevalence is much higher in some regions of Asia and South America 235.
Swyer–James syndrome (also known as McLeod syndrome) refers to a particular radiographic appearance, in which the disease predominantly affects one lung; the affected lung being smaller, probably as a result of poor lung growth 230. Histologically and clinically it does not differ from more generalised forms of PIBO.
Infection leading to PIBO typically affects pre-school children and results in a severe acute illness, usually with fever, respiratory distress and an oxygen requirement. Bronchopneumonia is common. Hospital treatment is nearly always required. Mechanical ventilation may be needed and is an independent risk factor for developing PIBO. PIBO is recognised clinically by persistent breathlessness, tachypnoea, hyperinflation, crackles and wheezing, usually with hypoxaemia for ≥30 days after the initial infection 229.
Lung function tests show severe airway obstruction that cannot be reversed by inhaled β-agonists 236. Pulmonary function in childhood PIBO can deteriorate over time as a consequence of the ongoing inflammatory process 230, 237.
Chest radiograph shows hyperinflation with or without areas of atelectasis and increased interstitial markings. Typical findings on high resolution computed tomography chest scan are patchy areas of hyperinflation giving a mosaic pattern of attenuation. This is best demonstrated by inspiratory and expiratory views, with air trapping becoming more apparent in expiration. There may also be bronchial dilation, bronchial wall thickening, areas of collapse and bronchiectasis 238.
The analysis of the cytological and immunophenotypical profile of bronchoalveolar lavage fluid shows an increase in neutrophils and lymphocytes with a predominance of CD8+ cells 237. There is a risk of exacerbation with subsequent respiratory tract infections. Long-term follow-up is required. The prognosis remains guarded and death following progressive respiratory failure has been reported.
Treatments that have been tried: anecdotal reports only
1) Bronchodilators: inhaled β2-agonists have been used for symptomatic treatment of wheeze. There is often a bronchodilator-responsive component to the fixed airway obstruction.
2) Inhaled corticosteroids: used because they might influence the underlying inflammatory process.
3) Systemic steroids (either oral prednisolone or i.v. pulsed methyl prednisolone): anecdotal evidence of efficacy in severely oxygen-dependent children.
4) Hydroxychloroquine.
5) i.v. immunoglobulin.
6) Antibiotics: used during infective exacerbations, particularly in children with coexisting bronchiectasis; azithromycin may have benefit as an anti-inflammatory agent as well as an antimicrobial.
7) Oxygen: should be used if hypoxaemia is present.
Treatment of PIBO: summary and recommendations
Side-effects
None of the treatments have been subjected to systematic assessment of side-effects specific to their use in PIBO.
Conclusions
PIBO is a heterogeneous disease syndrome based on presentation and histology. No proven effective treatment is available to date.
Prioritised research questions
1) Define exact aetiopathogenesis of PIBO.
2) Establish disease markers for differential diagnosis.
3) RCTs with large patient numbers and appropriate design.
PAEDIATRIC ILD
Treatment of paediatric ILD: the evidence base
ILD in children represents a very heterogeneous group of diseases with a variable natural history. Increasing knowledge of some genetic disorders causing ILD in children, such as surfactant protein deficiencies, is responsible for recent advances in the classification of these disorders. Moreover, the former classifications have not always been based on a systematic and independent analysis of the clinical phenotype, the genetic background, lung function and radiological data, and, most importantly, lung histopathology. A large European study classified ILD at all ages, but lacked a systematic pathological examination 239. An American study looked at children aged up to two years, with detailed independent pathological review 240, but their data on those aged >2 yrs have only been published as an abstract 241. Thus, there is no uniformly satisfactory classification spanning all ages in children.
We propose the constitution of an international (Europe, USA and Canada) registry with a prospective collection of cases, which should include data on precise clinical phenotype, genetic background, lung and radiological data, and histopathology of lung tissue, with all data being evaluated by independent experts in the field. Such a register has been started in France on a national basis by the group of A. Clement (Centre de référence des maladies respiratoires rares, CHU Hôpital d’Enfants Armand-Trousseau, Paris, France; the French national reference centre for rare diseases). To date, there have been no RCTs in children with ILD. Therefore, we can only extrapolate from adults (where the disease may be different) and report case series. The level of evidence is, thus, wholly unsatisfactory.
ILD of unknown origin
As lung parenchymal inflammation represents a common feature in these diseases, corticosteroids, such as oral prednisone or pulsed methylprednisolone are usually prescribed, but without any evidence-based recommendations. Case series of reports have shown that methylprednisolone may be effective, even when other steroids fail 242, 243. Hydroxychloroquine has been used in fibrotic changes on lung histology in some individuals 244–246. A favourable clinical outcome has been reported with aerosolised reduced glutathione in a child with chronic ILD 247.
ILD of known origin or associated with other diseases
ILD and connective tissue diseases
Hydroxychloroquine has been used in a 12-yr-old child with lymphocytic interstitial pneumonitis that preceded polyarticular rheumatoid factor positive juvenile arthritis 248. Other treatments include immunosuppressive agents, such as azathioprine, cyclophosphamide, cyclosporine (most often used in respiratory involvement of juvenile dermatomyositis 249) or methotrexate 244. All these immunosuppressants have been used in children with Wegener's granulomatosis 250. Anti-tumour necrosis factor (anti-TNF)-α therapy 251, 252, bronchoalveolar lavage with diluted surfactant 253, and monthly, high-dose i.v. immunoglobulin therapy have also been used with variable success 254. Cytosine arabinoside, vincristine and prednisolone, as well as doxorubicin and cyclophosphamide, have been used in patients with Langerhans cell granulomatosis and ILD 255.
Alveolar proteinosis
This condition is now classified as follows.
1) Surfactant protein deficiency: corticosteroids may enhance the surfactant protein B transcription and are thus generally used in clinical practice without any proven clinical benefit 256.
2) Granulocyte-macrophage colony-stimulating factor (GM-CSF) receptor issues.
3) GM-CSF auto-antibodies (largely adult disease): a successful treatment with inhaled GM-CSF has been reported in a child with anti-GM-CSF antibodies 257.
4) Macrophage blockade (lymphoma, etc.; largely adult disease): successful treatment of alveolar proteinosis with lung lavage has been reported in children 258–262.
Sarcoidosis
Because the cause of sarcoidosis is unknown, no specific therapy is available. Corticosteroids are widely used and are often effective. Methylprednisolone pulses have been able to improve lung function in patients with very severe ILD 263. Inhaled corticosteroids have been used for maintenance treatment in children but in an uncontrolled design and without being able to prevent relapses 264. Hydroxychloroquine therapy has been used with success in two Australian male children with diffuse pulmonary sarcoidosis 265. Successful steroid-sparing treatment of severe renal failure secondary to limited renal sarcoidosis was reported in a child treated with mycophenolate mofetil 266. Anti-TNF-α strategies have also been used in children with sarcoidosis 267, 268.
Eosinophilic pneumonia
A rapid beneficial response is usually observed after corticosteroids 269.
Idiopathic/infantile pulmonary haemosiderosis
Corticosteroids and hydroxychloroquine therapy are generally prescribed with variable efficacy 270–273. Treatment with low-dose cyclophosphamide and azathioprine has also been reported 271, 274, 275.
Desquamative interstitial pneumonia
A successful clinical response with hydroxychloroquine has been reported 276, 277.
Extrinsic allergic alveolitis
Avoidance of the allergen is mandatory. Oral or inhaled corticosteroids may be helpful 278–280, as may be pulsed methylprednisolone 281.
Other entities
Important ILDs which have recently been described include the surfactant protein gene defects (SpB, SpC and ABCA3), neuroendocrine cell hyperplasia of infancy (NEHI) and pulmonary interstitial glycogenosis 282. Surfactant protein deficiencies are a cause of considerable morbidity and mortality, and there is no known treatment. NEHI gets better without treatment, but the child may have prolonged oxygen dependency in the early years 283. Pulmonary interstitial glycogenosis generally improves with oral steroids 284. However, for all three conditions, there is an urgent need for new treatments.
Treatment of paediatric ILD: summary and recommendations
Side-effects
The commonest treatments used in paediatric ILD are corticosteroids and immunosuppressives. The side-effects of corticosteroids are hypertension, diabetes and bone demineralisation; for immunosuppressives, such as methotrexate, they are opportunistic infections and bone marrow failure. These patients need the usual careful monitoring.
Conclusions
ILD is a serious condition in childhood with significant mortality and morbidity. The treatments used have significant and sometimes severe side-effects yet there have been no RCTs to supply evidence to support anything we do. This is a situation in which a pan-European approach is an extremely good way of making progress.
Prioritised research questions
The constitution of an international register would allow: 1) an improved understanding of the epidemiology of ILD in children; 2) an analysis of phenotype–genotype correlations in genetic causes of ILD; and 3) an evaluation by RCTs of the potential treatments, including macrolides.
SUMMARY OF EVIDENCE BASE AND RECOMMENDATIONS FOR FURTHER STUDIES
The evidence base for considering the use of medicines in respiratory diseases only occurring in children is variable and sometimes absent. In the seven diseases discussed in this report, our knowledge of treatment benefits is virtually nonexistent in the management of congenital lung abnormalities and in PIBO, either as prevention or as treatment. The management of croup is now much more satisfactory and there seems little demand for new clinical trials. Now that ILD in children is better classified, there is a real chance we can begin to develop clinical trials to assess the efficacy and safety of treatment regimes. Because of the relative rarity of these disorders, studies will need to be developed on a European or even a global scale.
The evidence accrued to date and the ensuing knowledge gaps relating to studies needed in BPD, AVB and viral-induced wheeze can be summarised below.
1) To prevent the establishment of BPD, studies are needed to assess the value of iNO, to evaluate the usefulness of inositol and other nutrient supplements, and to investigate the benefit of various antioxidants.
2) To treat BPD once it has been diagnosed, there is value in assessing the effectiveness of surfactant therapy as well as investigating the role of various pulmonary vasodilators.
3) Until an effective vaccine is discovered to universally protect infants from developing AVB, we should focus on clarity and uniformity of definition of this viral infection, undertake separate studies for inpatient and outpatient sufferers due to differing disease severities and fully evaluate therapies such as the use of nebulised hypertonic saline. The potential effect of inhaled corticosteroids and leukotriene receptor antagonists on the development of recurrent respiratory symptoms should also be studied further.
4) Finally, in the more nebulous “viral-induced wheeze” classification, similar well-constructed RCTs with sufficient patient numbers need to be developed to examine the true benefit of high-dose inhaled corticosteroids and the use of leukotriene receptor antagonists in both acute viral wheeze and in recurrent viral wheeze phenotypes.
We must ensure some of these recommendations are taken up by study groups and funding organisations willing to support progression of our understanding of medicines that will have a major impact on the lives of our respiratory paediatric patients.
Statement of interest
Statements of interest for W. Lenney, A.L. Boner, L. Bont, A. Bush, K-H. Carlsen, A. Greenough, J. Kimpen and F.M. de Benedictis can be found at www.erj.ersjournals.com/misc/statements.dtl
Acknowledgments
The authors' affiliations are as follows. W. Lenney: Dept of Child Health, University Hospital of North Staffordshire and Keele University, Stoke-on-Trent; A. Bush: Dept of Paediatric Respirology, National Heart and Lung Institute, Royal Brompton and Imperial College, London; A. Greenough: Dept of Child Health, Kings College Hospital, London; J. Grigg: Academic Unit of Paediatrics, Institute of Cell and Molecular Science, Barts and The London Medical School, London; J. Hull: Dept of Paediatrics, John Radcliffe Hospital, Oxford (all UK). A.L. Boner: Dept of Paediatrics, University of Verona, Verona; F.M. de Benedictis: Dept of Paediatrics, Salesi Childrens Hospital, Ancona (both Italy). L. Bont: Dept of Paediatrics, University Medical Center, Utrecht; J. Kimpen: Wilhelmina Childrens Hospital, Utrecht (both The Netherlands). K-H. Carlsen: Dept of Paediatrics, Rikshospitalet University Hospital, University of Oslo, Oslo (Norway). E. Eber: Paediatric Dept, Medical University of Graz, Graz; M.H. Götz: Dept of Paediatrics and Adolescent Medicine, Wilhelminenhospital, Academic Teaching Hospital, Vienna (both Austria). B. Fauroux: Pneumologie Pédiatrique, Université Pierre et Marie Curie/Hopital Armand Trousseau, Paris (France). M. Sánchez Luna: UCIN, Hospital Materno-Infantil, Madrid (Spain).
F.M. de Benedictis and W. Lenney thank C. Ramacogi (Dept of Paediatrics, Salesi Hospital, Ancona, Italy) and C. Lenney (Nantwich, UK) for secretarial support.
- Received November 4, 2008.
- Accepted March 18, 2009.
- © ERS Journals Ltd
References
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
