Paediatrics in Vienna

The 2009 European Respiratory Society (ERS) Congress in Vienna, Austria, included a large paediatric programme, with many high-quality scientific presentations taking place. Herein we will present an overview of the highlights of the abstracts presented by the Paediatric Assembly. The abstracts were selected by the ERS Officers from each of the paediatric scientific groups and are set in the context of contemporary paediatric respiratory literature.

PAEDIATRIC ASTHMA AND ALLERGY

PAEDIATRIC RESPIRATORY EPIDEMIOLOGY

A number of old and new cohort studies have provided new insight into the aetiology, phenotypes and long-term outcome of childhood wheeze. In the Swedish Twin register, a 1,000 g decrease in birth weight was associated with an adjusted odds ratio for asthma of 1.57. Co-twin control analyses found an OR of 1.25 (95% CI 0.74–2.10) for dizygotic same-sexed, and 2.42 (95% CI 1.00–5.88) for monozygotic twins, with similar results for term and pre-term infants 32, 33. This study provides further evidence for an association between fetal growth and childhood asthma, independent of gestational age and of shared environment and genes.

Although the evidence supporting the hygiene hypothesis remains inconclusive 34, to date the strongest data have been shown for a potential protective effect of farm life 35. The mechanisms by which environmental exposures on a farm could confer protection from respiratory allergies are, however, not well understood. Data presented from the GABRIEL study supported a protective effect of raw unskimmed farm milk against asthma, hay fever and atopic sensitisation 36. Further analysis of the biological components of shop and farm milk showed higher concentrations of fat, lactoferrin and somatic cells in farm milk but, surprisingly, lower bacterial loads 36. Furthermore, the GABRIEL study provided novel evidence that farm life protects not only from respiratory symptoms, but also improves lung function 37 and decreases levels of exhaled nitric oxide 38.

There was quite some debate about the possible role of paracetamol in the pathogenesis of asthma 39, 40. A pilot study from the ALSPAC cohort explored whether DNA methylation could explain the association. Results remained inconclusive due to low numbers 41. In the Leicester cohort, a strong positive correlation was found between paracetamol intake and subsequent asthma. However, this effect largely disappeared after controlling for upper respiratory morbidity and reverse causation, suggesting that the association between paracetamol and asthma might be spurious rather than causal 42. This limitation does not apply to studies on paracetamol use during pregnancy. Only randomised trials can give the ultimate answer to this clinically relevant question.

A growing number of studies have begun to define different phenotypes of wheeze using multidimensional approaches to classification 43–45. Authors from the Paris birth cohort, analysing recurrent respiratory symptoms in infants, presented evidence for the existence of two different respiratory phenotypes in this age group 46. However, these multivariate methods will always find clusters, whether these really exist or not. To find a way out of these circular research paths, Swiss workers took an objective statistical approach to decide whether the variability within childhood wheeze is better explained by distinct asthma phenotypes (e.g. different diseases) or by severity gradients. The first results, based on cross-sectional symptom data from the Leicester cohorts, support a gradient rather than a class model 47, but the analysis must be extended to include physiological measurements and longitudinal data.

Although the age-related changes in the sex ratio for asthma have been known for decades, the mechanisms explaining these remain poorly understood 48, 49. An inverse relationship between testosterone levels and bronchial responsiveness was found in 321 young males 50, suggesting that sex hormones may help to explain the natural history of asthma through childhood.

CYSTIC FIBROSIS

Protocols for cystic fibrosis (CF) newborn screening usually utilise a stepwise approach with serum immunoreactive trypsin (IRT) as the initial test and genetic analysis for the most common mutations on samples with positive IRTs. Extended genetic testing platforms and different assays to assess pancreatic function, such as serum pancreatic associated protein (PAP), could potentially improve the diagnostic accuracy of newborn screening for CF. A study from the Netherlands showed similar positive and negative predictive values for the combination of IRT and PAP compared with a standard IRT/DNA programme 51. Extended genetic testing yielded additional screen positive subjects, but raises a clinical dilemma, since the clinical significance of most of the detected mutations is currently unclear. As genetic testing will also detect a significant proportion of unaffected carriers of gene mutations, protocols not requiring genetic testing would ultimately be preferable.

While newborn screening enables early initiation of treatment, two presentations from Australia documented early disease progression in infants despite diagnosis by newborn screening. The Arrest CF trial combines imaging techniques (computed tomography; CT) with assessments of lung function and bronchoscopy derived specimens to quantify infection and inflammation. This combined approach will shed more light on the complex relationship between these different aspects of CF lung disease in a longitudinal trial. In a preliminary cross-sectional analysis, the investigators assessed the link between ventilation distribution assessed by multiple breath washout and structural lung disease, and did not find a close link between the two methods 52. However, bronchial dilatation was found to be correlated with neutrophil elastase activity in BAL fluid raising the interesting question as to whether elastase activity may be a predictor of bronchiectasis 53. Longitudinal rather than cross-sectional data will help to better understand the relevance of these CT findings as there is limited information on the appropriate techniques to detect early bronchiectasis in this age group. Interestingly, the quality of magnetic resonance imaging (MRI) based imaging of CF lung disease is improving rapidly and could potentially replace CT as a radiation neutral technique in the future 54.

Early aggressive intervention is thought to be a major factor prolonging life expectancy in CF, but it is less clear whether treatment has been equally successful in improving the outcome of patients with advanced disease. A cohort study from London documented significant improvements in survival of patients with an FEV₁ <30% over the last 20 yrs 55. This highlights the need to continuously re-evaluate the criteria for lung transplantation which are largely based on a study performed almost 20 yrs ago that demonstrated FEV₁ <30 % to be associated with a 50% 2-yr mortality in CF patients 56.

Hypertonic saline (HS) has been shown to reduce pulmonary exacerbations and improve lung function in CF patients, and is currently being studied as an early intervention strategy in infants and young children. HS facilitates sputum expectoration, but also increases airway surface liquid fluid which is depleted in CF 57. An alternative approach is to use a sugar rather than a salt as an osmotic agent, and early studies have demonstrated that mannitol administered as a dry powder may have similar efficacy 58. A phase III trial of inhaled dry powder mannitol treatment demonstrated significant benefits in lung function 59. Effects on pulmonary exacerbations were less clear cut. Further studies should help to better understand the efficacy of this treatment 60.

PAEDIATRIC RESPIRATORY PHYSIOLOGY

Exciting new technology has recently been developed to assess lung growth in infants and young children 61. The apparent diffusion coefficient (ADC) of hyperpolarised helium 3 (³He) by MRI is a noninvasive, apparently robust method for assessing alveolar size 62. 49 term-born healthy children (25 male) aged 7–16 yrs (median 12.3yrs) underwent measurement of functional residual capacity (FRC) by plethysmography and ADC by MRI. ADC was approximately normally distributed, and was unrelated to age and FRC 61. These observations can only be explained by ongoing formation of new alveoli throughout childhood, which contradicts the current paradigm that alveoli only multiply for the first 3 yrs of life, and that subsequent lung growth is by alveolar enlargement. This is important because it signifies the potential of the lung to recover from early life insults. Conversely, it is possible that drugs, diseases and environmental exposures in later childhood may affect final structure and function of the alveoli.

There has been increasing attention over the last decade upon refining reference equations for lung function measurements in children. The greatest focus has understandably been on spirometry, but a presentation from Indianapolis (USA) described normative data for diffusion capacity of the lung for carbon monoxide (D_L,CO), with simultaneous measurement of haemoglobin (Hb) in 387 healthy individuals aged 6–18 yrs 63. Equations for D_L,CO, alveolar volume (V_A), D_L,CO/V_A and D_L,CO/Hb corrected for sex and Caucasian versus African–American race were presented. Two key points were that adjusting for Hb did not affect the result (as long as Hb is in normal range), and normalising D_L,CO for V_A did not provide a benefit 63. A six-centre study from Europe and South America analysed 1,117 measurements of interupter resistance (R_int) in children aged 2–13 yrs, which were used to develop sex-specific reference equations 64. A close relationship was observed between the data obtained in different centres suggesting that these reference values are consistent between populations.

Two studies described the effect of pre-term birth on lung function measured during preschool years. From a group in northern Italy, 74 children born at a mean gestational age of 29 weeks underwent R_int and impedance measurement from forced oscillation technique (FOT) 65. Results were comparable to those obtained in healthy children born at term. From Perth, Australia, a cohort of 150 children (74 bronchopulmonary dysplasia (BPD), 44 non-BPD and 32 controls) underwent spirometry and FOT between 4–8 yrs of age 66. There were significant differences (p<0.02) between pre-term children (BPD and non-BPD) and healthy subjects in FEV₁, forced expiratory flow at 25–75% of forced vital capacity and FOT reactance but not forced vital capacity or FOT resistance. On a related topic, investigators from the EPICure study 67 reported a wide range of lung function tests in 49 children born extremely pre-term and 52 controls recruited from across the UK and measured at a mean age of 11 yrs 68. Spirometry was better at discriminating between groups than D_L,CO, lung clearance index, or resistance or lung volumes measured in the plethysmograph. However, the authors cautioned that different investigations gave complementary information 68.

PAEDIATRIC RESPIRATORY INFECTIONS AND IMMUNOLOGY

A new diagnostic method, vibration response imaging (VRI), allows atelectasis and lung infiltrates to be visualised. In VRI, acoustic contact sensors detect vibration energy in the airways, which is processed into images by dedicated software. In a small proof-of-concept study in 12 patients, VRI could differentiate normal lungs from pneumonia and foreign body aspiration 69.

The use of galactomannan in BAL fluid as a marker for invasive pulmonary aspergillosis may reduce the need for open lung biopsy. In a retrospective study of 41 immunocompromised children, elevated galactomannan levels in BAL fluid showed a sensitivity and specificity of 82.4% and 87.5%, respectively. However, a negative test did not reduce the use of anti-fungal treatment 70.

In the treatment of bacterial lower respiratory infections, the frequent use of oxyminocephalosporins, a group of third generation cephalosporins, induces extended spectrum β-lactamase producing bacteria leading to a new pattern of bacterial resistance 71. In most children with severe community acquired pneumonia, symptoms resolve within 24 h of starting intravenous penicillin therapy 72. Given the risk of inducing new bacterial resistance when using new classes of broad spectrum antibiotics, this study highlights that treatment of pneumonia in children can, and probably should, be started with narrow spectrum antibiotics 72.

Several studies have shown that treatment according to current pleural empyema guidelines may not always lead to optimal results 73–76. In the cases presented, early surgical drainage was associated with risk of developing bronchopleural fistula, and conservative management often appeared successful 73–76. However, early surgical decortication of pleural empyema resulted in shorter hospitalisations and reduced morbidity. It remains unclear how these different interventions impact on the long-term outcome of pleural empyema.

Morbidity and mortality due to acute lower respiratory tract infections during the first 2 yrs of life are highest for children in Papua New Guinea. Laing and co-workers 77, 78 found that polymorphisms of genes associated with innate and adaptive immunity were associated with earlier and more severe respiratory infections in these patients. This finding may expand our knowledge as to which defects in the adaptive immune system can increase susceptibility for lower respiratory infections.

NEONATALOLOGY AND PAEDIATRIC INTENSIVE CARE

Clinical and demographic data on patients receiving noninvasive ventilation (NIV) are important for the planning of services and audit of performance. A recently established national register of children receiving NIV in eight accredited centres across France has now collected data about rare lung diseases in 82 patients, and recruitment will continue for the forthcoming years 79.

Increasing interest is being focused on the long-term respiratory outcome of premature birth 80. Increased levels of carbon monoxide in exhaled breath of neonates in the neonatal intensive care unit on the third and 14th day of life significantly improved the ability to predict subsequent BPD 81. However, the addition of carbon monoxide measurements only minimally increased the predictive levels (increasing the area under the curve from 94% to 97%), and the clinical relevance of such measurements remains uncertain. In a large study recording upper respiratory infections, hospital admissions and medication use annually during the first 3 yrs of life in 2,700 ex-premature babies (<32 weeks of gestation), children with BPD developed more symptoms than controls 82. However, despite the large numbers there were no clear predictors of respiratory morbidity, except the duration of oxygen requirement in the neonatal period 82. A Swiss group has studied a very early cohort of 13 infants (now adults with a mean age of 38 yrs) with “old” BPD, following up their lung function from infancy to adulthood. As this group have become older, abnormalities of small airways appear to have stabilised; although, they remained well below the normal range 83. Interestingly, between 30–40 yrs of age, residual volume/total lung capacity ratios increased from 26% to 39%. It remains difficult to know what effect there may be from a “survival bias” in this group as a number of the original cohort have died.

Fatal and near fatal asthma remains a rare but worrying problem in children. In an interesting study from Germany, a dramatic fall in both fatal and near fatal asthma between 1997 and 2008 was reported 84. This may be as a result of increased use of inhaled corticosteroids.

In a study of cardiovascular response to acute hypoxia, 13 children aged 8–12 yrs with BPD were subjected to oxygen levels of 12%. In children with BPD, oxygen saturation decreased from 99% to 91%, compared with 92% in term-born controls 85. Apparently, former BPD patients tolerate hypoxia well.

A study from Sweden showed high levels of the plasticiser di(2-ethylexyl)phtalate (DEHP) in the urine of children with tracheostomies 86. DEHP is a well-recognised toxin and endocrine modifier, especially in children. As yet the implications of this finding remain uncertain.

PAEDIATRIC BRONCHOLOGY

Although the stridor associated with laryngomalacia usually disappears during the first year of life, laryngomalacia is associated with other bronchial anomalies in 15% of cases and associated clinical symptoms related to gastro-oesophageal reflux and microaspiration may occur 87, 88.

Almost 70% of infants with persistent and therapy-resistant wheezing have tracheo- or bronchomalacia (either primary or secondary to vascular compression), tracheal or bronchial stenosis, or tracheo-oesophageal fistula (TEF) 89. The differential diagnosis of persistent wheezing in infants should also include airway narrowing due to lymph node compression in tuberculosis 90.

Fibreoptic bronchoscopy is an important diagnostic tool in the assessment of congenital airway and lung malformations. In patients with congenital lobar emphysema, for example, fibreoptic bronchoscopy may disclose bronchial valve mechanisms such as bronchial mucosal folds or bronchomalacia. In patients with bronchogenic cysts and cystic adenomatoid malformations, the most frequent endoscopic diagnosis is tracheomalacia or bronchomalacia 91. During endoscopic examination of the airways in children with TEF, a catheter can be introduced into the fistula before surgical repair. This procedure lets the surgeon know exactly where the fistula is located. In addition, fibreoptic bronchoscopy to deliver methacrylate glue to a TEF may be a reasonable alternative to operative closure 92.

In children, BAL may both have diagnostic and therapeutic purposes 93. Even in the absence of atopy, BAL fluid from children with asthma contains activated eosinophils even in the absence of atopy 94. Although controlled studies are lacking, it has been shown that whole lung lavage is an effective treatment for alveolar proteinosis 95. Unfortunately, this treatment may fail when adequate ventilation cannot be maintained during the lavage procedure. A new technique in young children was presented to overcome this problem. In this procedure, sequential whole lung lavage was performed with the use of an Arndt bronchial blocker device (Cook Medical, Limerick, Ireland) inserted through a 4.5-mm endotracheal tube, using a 2.2-mm bronchoscope for guidance, with the balloon inflated in a main bronchus, allowing the isolated lung to be lavaged while maintaining adequate ventilation of the other lung through the endotracheal tube 96.

Treatment of severe lower airway obstruction in children continues to evolve. Slide tracheoplasty was presented as an appropriate technique for correcting congenital long-segment tracheal stenosis with complete tracheal rings 97. Airway stenting has been used successfully in severe lower airway obstruction in adults, and may be an attractive alternative in children 98. Metal stents usually achieve airway patency and, in most cases, clinical improvement, whereas silicone stents seem less successful 98. However, severe and sometimes fatal complications occur in a considerable proportion of children. These include the formation of granulation tissue and airway migration 99, 100. More data from larger series of patients are needed to identify patients who may benefit from airways stenting.

Statement of interest

None declared.