European Respiratory Society
Paediatric Lung Function

This book offers a comprehensive review of the lung function techniques that are available today in paediatric pulmonology. This field is still developing rapidly and equipment and software can tell us more than ever about respiratory physiology in health and disease in children with various lung disorders. Experts from around the globe have contributed and provide a state-of-the-art review of the techniques, with a special focus on the clinical applications and usefulness in diagnosing and treating children with chronic lung disease.

  • European Respiratory Society Monographs
  1. Page vii
  2. Page viii
  3. Page ix
  4. Page 1
    Correspondence: P.D. Sly, Telethon Institute for Child Health Research, P.O. Box 855, West Perth, WA 6872, Australia. E-mail:

    Chronic respiratory disease has its origins in early life exposures that interfere with the normal growth and development of the respiratory system. Airway development is completed early in pregnancy but much of the alveolar development occurs after birth, making the developing lung vulnerable to both pre-natal and post-natal exposures. Lung function growth tracks along trajectories after birth and lung function at birth is a major determinant of lung function throughout childhood. Premature birth per se results in abnormal lungs and these changes are magnified by mechanical ventilation and supplemental oxygen therapy. Environmental exposures that may decrease lung growth include: environmental tobacco smoke; ambient air pollution, especially traffic-related pollutants; indoor air pollution, including exposure to household chemicals and bioaerosols; and viral lower respiratory infections in early life. Lung function growth continues throughout childhood reaching peak adult values earlier in girls than in boys. The rate of the normal decline in lung function, when measured by standard spirometry, can be accelerated by adverse environmental exposures, especially by tobacco smoking. An understanding of the normal processes of lung development, of the windows of susceptibility and of the adverse environmental exposures that may inhibit lung growth will allow a more thorough knowledge of the processes underlying the development of chronic respiratory diseases.

  5. Page 16
    Correspondence: P.N. Le Souëf, School of Paediatrics and Child Health, University of Western Australia, c/o Princess Margaret Hospital for Children, G.P.O. Box D184, Perth, Western Australia, 6840 Australia. E-mail:

    Lung growth and lung function in early life are influenced by both purely genetic mechanisms and interactions between genes and environmental factors. Genes directly influencing lung growth are not well understood, whereas genes that may influence lung function and interact with environmental agents have been more extensively studied. A large number of environmental toxic agents have been implicated as having the potential to adversely affect normal lung development. As lung development occurs over such a long time, there are many windows of opportunity for a variety of environmental agents to exert their potential effect. There is strong evidence that in utero exposure to environmental agents, specifically tobacco smoke, is associated with adverse respiratory outcome and this appears to be influenced by genetic makeup. Multiple atmospheric pollutants have been associated with decreased lung function, increased hospitalisation with asthma and triggering of asthma exacerbations in children. Exposure to farm animals, pets and infections have also been demonstrated to influence respiratory outcome. In spite of the current knowledge base on the subject, a comprehensive understanding of the genetic and environmental factors influencing lung growth and lung function in early life is lacking.

  6. Page 35
    1. Page 35
      Correspondence: K-H. Carlsen, Oslo University Hospital, Dept of Paediatrics, NO-0027 Oslo, Norway. E-mail:

      Techniques using tidal breathing measurements have been employed in the assessment of lung function in infants and preschool children. These measurements have been used for diagnostic purposes, for prediction or used as risk factors for later respiratory disease and to evaluate lung development during the first year of life, including the effect of bronchodilating medication on the tidal breathing measurements. Standardisation of measurements and establishment of reference values is required, as well as a discussion about their interpretation, and clinical and scientific value.

      Quiet breathing is a condition for performing noninterventional tidal breathing measurements, which may then be obtained regardless of respiratory or arousal states. Measurements may be performed in both in- and outpatient settings, but are dependent upon a carefully trained staff.

      Tidal breathing may be measured either by tidal flow and/or volume measures obtained through the use of a pneumotachograph or an ultrasonic flow-sensor through a facemask, or through respiratory inductive plethysmography by use of bands around the chest and abdomen. Most studies have applied facemask and pneumotachographs, but ultrasonic flow meters have also recently become available for such measurements.

      With respect to clinical value, several studies have demonstrated tidal flow volume (TFV) loops to be valuable in screening and diagnosing upper from lower airways diseases by describing typical shapes patterns of the flow–volume (or flow–time traces) with bronchial and upper respiratory tract obstruction. With respect to bronchial obstruction, the most commonly employed parameter is the ratio of time to reach peak expiratory flow to total expiratory time (tPTEF/tE). Criteria for describing bronchial obstruction have been described by use of this parameter. Reduced lung function early in life (either neonatally or in early infancy) by using tPTEF/tE obtained by tidal breathing techniques has also been shown to be important in epidemiological studies of later development of recurrent wheeze or asthma, although not on an individual level.

      In common with several other lung function measurement techniques, it has been difficult to fully understand the complex physiological background of TFV measurements. Both respiratory mechanics and control of breathing are probably reflected in TFV measurements.

      One of the main advantages of tidal breathing measurements is the possibility of obtaining objective measures of lung function in diseased infants and young children without the need for sedation. Lung function by tidal breathing may be one of several objective measures for aiding diagnosis, monitoring disease and providing a research tool in both in- and outpatient settings.

    2. Page 46
      Correspondence: S. Lum, Portex Respiratory Unit, Respiratory Physiology and Medicine, University College London, Institute of Child Health, 30 Guilford St, London, WC1N 1EH, UK. E-mail:

      Forced expiratory manoeuvres can now be applied from birth throughout childhood, and continuous reference equations for white subjects of European descent are available from 3–80 yrs with which to characterise normal lung growth, assess the nature and severity of disease and monitor disease progression or resolution in both clinical and research settings.

      Outcomes derived from raised volume forced expiratory flow volume manoeuvres have proved to be among the most sensitive lung function tests for use in infants, but are relatively invasive, require sedation of the infant and considerable expertise during data collection and analysis, such that their use is generally limited to the research setting. Furthermore, published reference data for either the tidal or raised volume rapid thoraco-abdominal compression may not be appropriate for infant data collected with the current generation of commercially available equipment.

      There is an urgent need to develop appropriate quality control and over-read criteria for spirometry for both preschool and school-age children, and to extend spirometry reference ranges for use in all subjects, irrespective of ethnic origin. Despite its feasibility in children as young as 3 yrs and its widespread use for assessing airway function in both clinical and research settings particularly with respect to long-term impact of early life events, spirometry may be relatively insensitive to early lung disease, especially if outcomes are limited to forced expiratory volume in 1 s such that alternative lung function techniques may be required when determining the nature and magnitude of lung disease in early life.

    3. Page 66
      Correspondence: C.S. Beardsmore, Dept of Infection, Immunity and Inflammation (Child Health), University of Leicester, Robert Kilpatrick Clinical Sciences Building, Leicester Royal Infirmary, P.O. Box 65, Leicester, LE2 7LX, UK. E-mail:

      Plethysmography has been established for many years and is widely used in adult respiratory laboratories. Its application to infants is less common, because of the need for sedation and requirement for specialised equipment and highly trained staff. Recent developments in technology and computing have resulted in commercially available plethysmographs that are simple to operate. There is still a need for the operators to understand the physiological basis of the measurements, the assumptions behind them, and the technical aspects of the equipment. Without this understanding, errors can be introduced, ranging from using a chamber with a leak that is too short to accepting data from a subject who has not adhered closely to their instructions. Even when measurements are made by knowledgeable and experienced operators, there is a need for careful inspection and review of data.

      The measurements of resting lung volume via plethysmography (FRCp), and other divisions of lung volume (total lung capacity and residual volume), are well established in adults, children of school age, and (to a lesser extent) in infants. Fewer studies have focused on the preschool child but, with careful coaching, children as young as 3 yrs can be tested. Indices of airways resistance (Raw) or specific airways resistance (sRaw) can be measured with minimal cooperation, and without the need for a period of airway occlusion. Modern systems can compensate for the effects of heating and cooling of the respired air, but the accuracy of the compensation when compared directly with a system in which the subject breathes heated humidified air is still unclear. Work also remains to be done to establish the most appropriate indices of sRaw, and to standardise the protocols for data collection and analysis, but the measurement seems set for resurgence.

    4. Page 87
      Correspondence: P.D. Robinson, The Children's Hospital at Westmead, Locked Bag 4001, Westmead, NSW, 2145, Australia. E-mail:

      The multiple-breath washout (MBW) test is a patient-friendly noninvasive tool, which assesses the uniformity of ventilation distribution in the lungs and small airway function by recording the clearance of an inert marker gas during tidal breathing. Feasibility has been demonstrated from infancy through to adulthood. The lung clearance index (LCI), a widely reported parameter of overall ventilation inhomogeneity, has been shown to have remarkable agreement across wide age ranges of healthy subjects, between different equipment setups and different centres. In heterogeneously distributed disease processes, such as cystic fibrosis, which involve the small airways early on in their course, the LCI has been shown to be more sensitive than conventionally used lung function tests, and to correlate better with structural lung damage. New approaches to the analysis of washout curves, such as normalised phase III slope analysis, provides further insights into the location of the disease process within the small airways, and evolution of disease. Other disease conditions that may benefit from wider use of MBW include asthma and bronchiolitis obliterans. In asthma, MBW could improve our understanding of the underlying pathophysiology, in particular bronchial hyperresponsiveness, and provide an additional tool to assess the response to therapeutic interventions. Greater sensitivity provides the opportunity for earlier diagnosis and intervention, with potential improvement in respiratory outcome. Further studies are required to define clinically meaningful thresholds and there is a need for affordable commercial equipment that can be used in the clinical setting across the paediatric age range.

    5. Page 105
      Correspondence: N. Beydon, Functional Unit of Paediatric Pulmonology, APHP-Robert Debré Hospital, 48 Boulevard Sérurier, 75019, Paris, France. E-mail:

      Techniques based on airflow interruption to assess respiratory mechanics have in common the following: 1) they are noninvasive and easy to perform in young children and in infants; 2) they have been the subject of recent published recommendations on technical aspects and data acquisition; 3) the use of an apparatus, bacterial filter and mouthpiece with small deadspace and low resistance is of particular importance in young children and infants; and 4) there is a lack of clinical studies in large populations of healthy and sick children to clearly establish clinical applications, despite encouraging results in small groups.

      Single and multiple occlusion techniques (SOT and MOT) are suitable for respiratory compliance measurement in sleeping infants. In addition, the expiratory time constant and the respiratory resistance can be measured using SOT. Measurements can be performed at baseline with good intra-measurement, inter-observer and intra-observer variability. The occlusion techniques have been used to assess lung function in pre-term infants.

      Resistance measured with the interrupter technique (Rint) is more suitable for pre-school and school-age children unable to perform reliable flow–volume curves. Rint is so simple to implement that it can be used in the field. Normative values and short- and long-term repeatability data are available. Measurements after bronchodilator administration give more information on lung function than a single baseline measurement, but more data on longitudinal baseline measurements are needed. It is possible that Rint may be a suitable way to monitor lung function during clinical trials, and studies on large populations are required.

    6. Page 121
      Correspondence: F. Marchal, Service d'explorations fonctionnelles pédiatriques, Hôpital d'enfants, Allée du Morvan, 54500 Vandoeuvre, France. E-mail:

      The forced oscillation technique (FOT) allows the characterisation of respiratory mechanics such as respiratory resistance (Rrs) and reactance (Xrs). The primary advantage of the FOT is that little active cooperation is needed from the subject and measurements are easily performed during tidal breathing with firm support of cheeks and mouth floor. Standards are available for equipment specifications and data collection. Within-session variability should be documented, while between-session repeatability is essential for the study of bronchomotor responses.

      Most clinically relevant information has been gained from Rrs and/or Xrs in the frequency range 5–8 Hz, where Rrs is mostly dependent on frictional losses in the airways and Xrs is determined in great part by elastic respiratory properties. There has been an increasing number of paediatric reports during the past decade, particularly in uncooperative young children. Case–control studies have reported lung function abnormalities in young children in chronic lung disease of prematurity, with less consistent data in cystic fibrosis. Findings in children with stable asthma appear dependent on inclusion criteria, with more abnormalities detected with recruitment from asthma clinics than in field studies. Longitudinal studies in asthmatic children over weeks to months suggest the FOT and in particular Xrs is sensitive to anti-inflammatory management, including inhaled steroids or allergen avoidance. An acute >30% decrease in Rrs after bronchodilator inhalation is suggestive of asthma but a 15–20% decrease is not unusual in healthy children. The increase in Rrs and decrease in Xrs are consistent with spirometry in response to methacholine or histamine inhalation and also indicative of positive responses to indirect airway challenge such as inhaled adenosine 5-monophosphate, exercise or cold airway hyperventilation.

      The frequency response and time course of respiratory impedance have been helpful in understanding pathophysiological mechanisms involving lung visco-elasticity, ventilation inhomogeneity or airway compliance, and heightened bronchomotor tone in relation to volume history, in asthma and other obstructive airway diseases.

    7. Page 137
      Correspondence: M.W.H. Pijnenburg, P.O. Box 2060, 3000 CB Rotterdam, The Netherlands. E-mail:

      The fraction of nitric oxide in exhaled air (FeNO) is a marker of eosinophilic airways inflammation. The gold standard for measuring FeNO is the single-breath on-line method, which can be performed in children starting from the age of 4–5 yrs. Alternatively, FeNO can be measured off-line, preferably with a flow-controlled technique. In preschool children, FeNO can be measured on-line during tidal breathing or with a modified single-breath technique, with the expiratory flow controlled by the operator.

      For FeNO measurements, a chemiluminescence-based analyser or a recently introduced portable analyser using an electrochemical sensor can be used.

      FeNO measurements are increasingly used in paediatric respiratory diseases, in particular in the management of paediatric asthma. FeNO provides the clinician with information regarding underlying airway inflammation, complementary to symptoms, lung function testing and bronchoprovocation tests. FeNO may be helpful in diagnosing asthma, selecting patients who will benefit from inhaled corticosteroids and patients who require additional therapy from those whose medication dose could be reduced or withdrawn. However, dose titration studies show no or only modest effects on outcomes like symptoms and exacerbations. There are some indications that selected phenotypes of asthmatic patients might benefit most from FeNO monitoring.

      In other childhood respiratory diseases, FeNO measurements are of very limited value, except for primary ciliary dyskinesia, where nasal nitric oxide measurements might become the screening test of first choice.

    8. Page 155
      Correspondence: A. Moeller, University Children's Hospital Zurich, Division of Respiratory Medicine, Steinwiesstrasse 75, 8032 Zurich, Switzerland. E-mail:

      The analysis of exhaled breath condensate (EBC) in children is a challenging new technique and has potential as a noninvasive measure to assess and monitor airway inflammation, oxidative stress and metabolic processes. The technique is safe, simple to perform and ideally suited for young children and for repeated use in longitudinal studies.

      Various collector systems have been described so far, all with the same principle of cooling the exhaled air with consequent condensation of water vapour as well as impacting aerosol particles to the cold surfaces. There is striking variability of results between different studies and laboratories. There are a number of important reasons for this variability. The concentration of most of the biomarkers in EBC is at the detection limit of the assays and there is a high risk of contamination of the EBC contents by the upper airways and from ambient sources. There is evidence that condenser coating surfaces and the temperature at which EBC is collected influence the biomarker concentration to a variable extent. An unresolved issue is the highly variable dilution of nonvolatile substances within the EBC, not only between subjects but also within a single subject over time. Several attempts have been made to overcome this important issue, but the practical difficulties of measuring both a biomarker and a dilution marker within the relatively small volume of EBC have limited widespread use of dilution correction. Many different biomarkers have been measured in EBC obtained from children with different respiratory diseases, most in a cross-sectional manner, with the aim of describing markers that allow distinction between health and disease and between different entities of respiratory disease. In this chapter, we provide a detailed update on the current knowledge of EBC contents and discuss the available paediatric data.

    9. Page 183
      Correspondence: G.J. Hutten, Dept of Paediatric Respiratory Medicine, Academic Medical Centre (G8-211), Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. E-mail:

      Much of the burden of respiratory diseases in childhood and later life has its origins in infancy and early childhood. The largest groups of infants with respiratory problems are infants with wheezing disorders.

      Lung function measurements can be used to monitor the progression of the different wheezing phenotypes, and may be used to monitor the effect of possible therapeutic interventions, such as bronchodilators. Accurate assessment of lung function in infants is difficult to perform, is time and manpower consuming and sedation is often needed in older infants.

      In addition to lung function, measuring the electrical activity of the respiratory muscles may help to assess the temporal dynamic interaction of intercostal and diaphragmatic muscle activity with the resulting flow pattern as well as lung volume. A good feasibility and repeatability of transcutaneous electromyography of the respiratory muscles (tc-rEMG) in matched comparison to flow and lung volume in infants has been established. Combining noninvasive rEMG and lung function measurements may help to provide a more comprehensive picture of lung mechanics and its adaptive reaction in disease.

      In infants with bronchopulmonary disease, the temporal dynamic interaction between respiratory muscle activity, resulting flow and lung volume is altered in comparison with healthy counterparts. In infants with wheeze, the dynamic interaction improves after administration of β2-agonist, although tidal volumes and functional residual capacity remain similar. The temporal relationship of tc-rEMG to flow and the increase or decrease of adaptive variability provide additional information on adaptive mechanisms in infants with impaired lung mechanics which is not obvious in tidal breathing and lung volume measurements alone.

    10. Page 195
      Correspondence: T. Riedel, Paediatric and Neonatal Intensive Care, University Children's Hospital, Inselspital, CH-3010 Bern, Switzerland. E-mail:

      Electrical impedance tomography (EIT) is a noninvasive, radiation-free imaging technology able to detect regional differences in lung ventilation and to some extent also lung perfusion. Multiple studies for the validation of EIT have been performed, showing excellent results with respect to regional ventilation. Most of the research performed with EIT has focused on the critical care environment, especially on the optimisation of mechanical ventilation in patients with acute lung injury. The few publications documenting the use of EIT in spontaneously breathing subjects mainly describe neonates and infants, some adults and to date no children.

      Some technical limitations of the currently used EIT systems still have to be improved: signal-to-noise ratio, electrode positioning, etc. Despite more than 20 yrs of research on EIT, it is only recently that a group of experts have suggested an improved, unified reconstruction algorithm that now needs to be validated. Consensus is also needed for the definition of clinically relevant EIT parameters, normative values, standardised measurement conditions, filtering techniques and definition of regions of interest. Besides the technical issues and the development of critical care applications, future research needs to address the implementation of EIT in spontaneously breathing subjects, ideally in combination with commonly used and validated lung function tools.

      EIT is a very promising, relatively cheap, easily applicable lung function monitoring tool. Before being ready for clinical use, research needs to focus on standardisation of data acquisition and analysis and normative values.

    1. Page 209
      Correspondence: J. Hammer, Division of Intensive Care and Pulmonology, University Children's Hospital Basel, Römergasse 8, 4005 Basel, Switzerland. E-mail:

      Forced spirometry (e.g. forced expiratory volume in 1 s) performed in the context of bronchodilation and/or bronchial hyperreactivity testing is the best validated and most common pulmonary function tool used to diagnose and assess asthma in older children. However, spirometric measurements are often in the normal range in the majority of symptom-free asthmatic children (independent of disease severity or treatment), and their value for the short- and long-term evaluation of disease activity and airway remodelling remains controversial. Other pulmonary function tests, such as measures of airway inflammation, have yet to prove that they are more sensitive and helpful to guide anti-inflammatory treatment.

      Wheezing disorders in infancy are common, and their prevalence has increased in the last decade. Clinically, it is often difficult to distinguish recurrent episodes of transient “viral wheeze” from early asthma, which has implications for treatment and prognosis. There is a growing number of techniques available to assess pulmonary function in younger children, an uncooperative age group. Although these innovative techniques have increased our understanding of wheeze in infancy and early childhood, they have remained research tools and provide little help in the individual management of wheezing disorders in infants and toddlers.

    2. Page 225
      Correspondence: S.C. Ranganathan, Brighton and Sussex Medical School, 10th Floor Royal Alexandra Hospital for Children, Eastern Road, Brighton, BN2 5BE, UK. E-mail:

      Data suggesting that pulmonary disease begins early in those with cystic fibrosis (CF) highlights the importance of earlier, aggressive interventions to limit or prevent lung damage. Objective, sensitive and accurate outcome measures for such interventions are required. Infant and preschool lung function testing is at the point of being introduced into clinical practice for management of children with CF, influenced by data from recent studies, efforts to standardise equipment and the development of commercial equipment. In the absence of ideal tests, combinations of the best available tests should be considered. However, our ability to demonstrate that introduction of lung function testing in infants and preschool children with CF can improve both clinical management and prognosis depends on high-quality future studies in specific targeted areas.

    3. Page 240
      Correspondence: B. Fauroux, AP-HP, Hôpital Armand Trousseau, Paediatric Pulmonary Dept, Research Unit INSERM UMR S-938, National Reference Centre for Rare Lung Diseases, Université Pierre et Marie Curie-Paris 6, 28 avenue du Docteur Arnold Netter, Paris, F-75012, France. E-mail:

      The control of breathing is a complex and multi-factorial process, the aim of which is to maintain appropriate homeostasis for the arterial blood gases and hence for oxygen delivery to the tissues.

      The respiratory control system is composed of three functional components: the sensory receptors, which include the chemoreceptors and mechanoreceptors that provide information about the status of the respiratory system, the central integrating circuits, and the motor output to the respiratory muscles.

      The tests exploring the control of breathing are generally stimulus response tests, in which a receptor is stimulated and the motor output is measured, either during wakefulness or sleep. The measurements of the occlusion pressure and the ventilatory response to hypoxia or hypercapnia are tests that explore the control of breathing system. These tests, however, are poorly able to separate the three functional components of the control system and need to be interpreted within the patient's clinical context. In clinical practice, polysomnography seems to be the most informative test for the diagnosis, the prognosis and the treatment of a condition associated with an abnormality in the control of breathing system.

    4. Page 251
      Correspondence: E.J.L.E. Vrijlandt, Dept of Paediatric Pulmonology and Allergy, Beatrix Children's Hospital/University Medical Center Groningen, P.O. Box 30.001, NL 9700 RB Groningen, The Netherlands. E-mail:

      The pathophysiology of bronchopulmonary dysplasia (BPD) has changed during recent years. Classic BPD is characterised by fibrosis and smooth muscle augmentation of medium-sized airways. “New” BPD is dominated by interrupted acinar development and dysmorphic vasculogenesis. Infants and children with BPD: 1) continue to suffer from comparable degrees of combined obstructive and restrictive lung disease; 2) frequently have clinical evidence of airway hyperresponsiveness. Challenge tests provide valuable information, but these measurements are more complicated, which reduces their feasibility in younger children and precludes application of these techniques in older children only; 3) have lower pulmonary diffusing capacity than term controls; and 4) more often have exercise-induced bronchospasm and evidence for limitations in exercise capacity.

      While functional residual capacity may remain reduced in established BPD, it most commonly becomes normalised or moderately elevated due to hyperinflation, with or without air trapping secondary to airway obstruction. Indices of ventilation inhomogeneity may be less useful when assessing the effect of prematurity or lung disease in neonates. In conclusion, to fully understand the impact of immaturity on the developing lung, it is unlikely that a single parameter will accurately describe underlying changes. Therefore a combination of lung function tests is recommended to fully characterise the individual patient with BPD.

    5. Page 263
      Correspondence: E. Eber, Klinische Abteilung für Pulmonologie und Allergologie, Univ.-Klinik für Kinder- und Jugendheilkunde, Medizinische Universität Graz, Auenbruggerplatz 30, 8036 Graz, Austria. E-mail:

      Congenital abnormalities of the lung, individually infrequent but as a group important, comprise a wide range of abnormalities of the bronchial, arterial, venous and lymphatic trees. In addition, abnormalities of the chest wall (including the diaphragm) are of relevance in this context. In this chapter, the most prominent abnormalities of the lower respiratory tract, i.e. tracheo- and bronchomalacia, tracheal and bronchial stenosis, tracheo-oesophageal fistula, the so-called cystic malformations of the lung, and congenital diaphragmatic hernia, are discussed.

      Lung function tests able to detect changes in flow limitation may be very useful in children with suspected or proven central airway obstruction. In infants and young children, tidal flow–volume loop analysis and forced expiratory flow–volume curves may be valuable as a screening and diagnostic tool; they may allow for separating extra- from intrathoracic airway diseases, monitoring of disease progression, and assessing the effects of therapeutic interventions. Analogous to forced expiratory flow–volume curves in noncooperative children, maximal expiratory flow–volume curves generated by cooperative children are an established tool to delineate and quantify large airway obstruction. In addition, the assessment of airway resistance may contribute to clinical decision making in this situation. Measurements of static lung volumes by plethysmography and/or gas dilution techniques are crucial for understanding lung growth and development, particularly in patients with space-occupying abnormalities and chest wall defects. The multiple-breath inert gas washout technique, requiring only passive cooperation and tidal breathing, can be performed in subjects from any age group and appears to be a very promising method for the early detection and quantification of abnormal ventilation distribution in the peripheral airways resulting from airway obstruction or abnormal branching.

    6. Page 277
      Correspondence: A. Schibler, Paediatric Intensive Care Unit, Mater Children's Hospital, South Brisbane 4101 QLD, Australia. E-mail:

      Recent studies have highlighted the need for careful titration of ventilatory support to minimise ventilator-induced lung injury. To achieve this, the measurement of both respiratory mechanics and gas distribution in the lung is required. Measurement of dynamic compliance during recruitment can identify an optimum positive end-expiratory level. Pressure–volume curves measure recruitable lung and characterise quasi-static compliance of the respiratory system. Intrapulmonary gas distribution and regional lung mechanics can be measured on a breath-by-breath basis using electrical impedance tomography.

    7. Page 291
      Correspondence: N. Regamey, Division of Paediatric Respiratory Medicine, Dept of Paediatrics, University Children's Hospital, 3010 Bern, Switzerland. E-mail:

      Spirometry and other lung function tests can be used to diagnose and guide therapy in children with lung diseases. However, they only assess resting lung function indices. Cardiopulmonary exercise testing (CPET) allows evaluation of overall exercise capacity, both in healthy children and in those with chronic disease. During exercise testing, functional deficits that were not apparent during conventional pulmonary function testing may be identified.

      Indications for exercise testing include assessment of exercise capacity, detection of adverse reactions to exercise, such as exercise-induced bronchoconstriction, arrhythmia or hypoxia, and evaluation of the functional impact of chronic illnesses on children. The results of exercise tests can be used as a guide to prescribing safe and individual exercise programmes. They can also provide confidence to the child, caregivers, teachers and primary care physicians that it is safe for the child to exercise.

      Maximal aerobic exercise capacity can be determined directly by the measurement of maximal or peak oxygen uptake during formal CPET on a treadmill or a cycle ergometer. It can also be assessed indirectly by measuring heart rate or power output during maximal or submaximal exercise, and can be estimated in field tests. Field tests are simple to perform and inexpensive and are often used in epidemiological surveys, in the rehabilitation setting or in advanced disease. Their main limitation is that they are difficult to standardise. Formal CPET allows assessment of various ventilation parameters such as oxygen uptake, ventilatory anaerobic threshold and flow limitation. It is particularly suited to determine the aetiology of impaired exercise tolerance.

      There is no “best” exercise test as such. Whichever test is preferred depends not only on the indication and the clinical question, but also on the condition of the subject tested, the experience of the tester with a specific protocol, safety issues and financial resources.

    8. Page 310
      Correspondence: C. Thamrin, Division of Paediatric Respiratory Medicine, University Children's Hospital of Bern, Inselspital, 3010 Bern, Switzerland. E-mail:

      Studying lung function variability over time forms part of an increasingly advocated approach to treat living systems in health and disease as real, nonlinear complex dynamic systems, instead of as idealised, linear components. Detrended fluctuation analysis (DFA) is a technique that characterises the fractal behaviour of fluctuations in lung function over time, while taking into account the nonstationarities often seen in physiological signals. DFA has been used to demonstrate both the presence of and alterations to fractal behaviour in respiratory measures during infancy, ageing and different disease states, and there is some evidence that it can provide prognostic information of clinical benefit. Due to its widespread use, DFA has been extensively validated in different physiological systems, and its limitations reasonably well quantified.

      In the lung, fractal behaviour may arise out of the fractal structure of airway branching, intrinsic neurological control or external environmental fluctuations. Fractal behaviour is thought to facilitate error tolerance and adaptability, and deviations from this behaviour may have implications for disease characterisation and detection. What remains to be shown is whether there is indeed a “normal”, healthy level of variability in physiological signals, reflective of an ability to adapt that lies between randomness and complete determinism, and whether fractality in respiratory measures would correspond with other measures of complexity to build a more complete picture. Also important is more information on how this is altered during development and disease, particularly in children, and whether these observations would indeed provide clinical value in predicting disease.