Wheeze is a common symptom in young children and is usually associated with viral illnesses. It is a major source of morbidity and is responsible for a high consumption of healthcare and economic resources worldwide. A few children have a condition resembling classical asthma. Rarer specific conditions may have a wheezy component and should be considered in the differential diagnosis. Over the last half century, there have been many circular discussions about the best way of managing preschool wheeze. In general, intermittent wheezing should be treated with intermittent bronchodilator therapy, and a controller therapy should be prescribed for a young child with recurrent wheezing only if positively indicated, and only then if carefully monitored for efficacy. Good multidisciplinary support, attention to environmental exposition and education are essential in managing this common condition. This article analyses the pathophysiological basis of wheezing in infancy and critically discusses the evolution of the scientific progress over time in this unique field of respiratory medicine.
- preschool children
Statistics from Altmetric.com
When the facts change, I change my mind. What do you do, sir?
—John Maynard Keynes, economist
Two extensive reviews1 ,2 and a European Consensus Statement3 on classification and management of wheezing in preschool children have recently been published. This article aims to summarise the evolution of the concepts behind these manuscripts and the intellectual progress over time in one of the most fascinating fields of paediatric respiratory medicine in order to see what lessons can be learned from the ebb and flow of past ideas.
What is wheeze and where does it come from?
Wheeze is an onomatopoeic, musical sound generated by bronchial wall vibrations that occurs when respiratory effort exceeds that required to achieve maximum airflow within the airway. It is typically a high-pitched, expiratory sound associated with increased work of breathing and can also sometimes be heard in inspiration.
Wheeze indicates partial airway obstruction, for example, due to bronchospasm, intraluminal secretions, inflammation or other structural changes in the airway walls, or to dynamic compression of the airway. Wheeze associated with distal airway disease originates from the larger bronchi rather than from the affected airways, and this is more likely due to compression of the larger airways secondary to positive pleural pressures generated during expiration to overcome increased airway resistance.
There are several reasons for wheeze being frequent in infants and young children. First, the trachea and major bronchi are more compliant in this age range. Second, peripheral airway resistance is disproportionately high in infants. Third, the relative lack of the elastic recoil of the lung in infancy predisposes to early airway closure, even during tidal breathing. Finally, a compliant ribcage and the mechanically disadvantaged diaphragm in young children are both responsible for the increased severity of respiratory symptoms at this age.4
Wheeze in young children is common. Indeed, it is reported by about 30% during the first three years of life.5 The most common clinical phenotype in young children is episodic viral wheeze, a disorder characterised by acute episodes of wheeze, cough and breathlessness in association with a viral respiratory tract infection, with few or no interval symptoms. Episodes may be severe enough to warrant hospital admission, but the illness is generally mild and usually resolves with age.6 ,7 A few children have a condition resembling classical asthma (multiple trigger wheeze is the preferred term), with interval symptoms between viral-induced episodes, usually precipitated by cold air, activity, laughing or crying, and often with an atopic personal and family history. Rarer specific conditions, such as congenital airway disorders, cystic fibrosis, primary ciliary dyskinesia, chronic lung disease of prematurity and obliterative bronchiolitis, may have a wheezy component. Overlap or the coincidental occurrence of two independent disorders may occur in individual children.
Interestingly, some of the infants who wheeze appear to be unperturbed by their respiratory symptoms despite laboured breathing. Wheeze is most obvious when the child is active and playing, and is usually inaudible when asleep. These happy wheezers are lively and cheerful, have good appetite, develop normally and are not distressed at all by their wheeze.4 It has been suggested that these infants need not to be treated. In such a setting, however, parents may be psychologically disturbed by hearing their child wheeze, physically troubled by sleepless nights, and are typically unhappy to accept the idea that nothing can be done to relieve this respiratory noise.
Nanos gigantium humeris insidentes (dwarfs standing on the shoulders of giants)
The 12th-century theologian and author John of Salisbury used a version of this phrase in his treatise on logic, written in Latin in 1159. He pointed out that we see more and farther than our predecessors, not because we have keener vision or greater height, but because we are lifted up and borne aloft on their gigantic stature.8
Over the last 50 years, paediatric pulmonologists spent much of their time trying to resolve whether preschool wheeze is a single disorder or encapsulates distinct phenotypes. In 1969, Williams and McNicol9 first speculated that wheeze associated with respiratory viral infection (namely wheezy bronchitis) could not be separated from asthmatic wheeze. They proposed that there was a broad spectrum of wheezing conditions in young children, at one end of the spectrum being children with ‘mild and evanescent’ wheeze that was resolved by school age, and at the other end being children with recurrent episodes of wheeze from early life that continued into school age. This approach wrongly conflated duration and severity of symptoms with their temporal pattern, a cardinal error still prevalent today. The consequence of this conceptual assumption—and treatment accordingly—was the assertion that “any infant with recurrent episodes of wheezing should be considered as having asthma, regardless of age of onset, evidence of atopy, apparent precipitating cause of wheeze, or frequency of the wheeze”.4 However, a number of clinical and epidemiological studies supported the evidence that these two disorders are totally independent.10 ,11 Denial of the former ‘truth’ was highlighted by these unequivocal words: “The fact that these conditions were lumped as a single disorder is testament to the arrogance of hospital doctors and to the lack of respect afforded to clinical epidemiologists.”12
Since then, several approaches have been used to categorise preschool wheezers: epidemiological (namely ‘transient early wheeze’ vs ‘persistent wheeze’5 and various more sophisticated classifications);13–16 symptom pattern (namely ‘episodic viral wheeze’ vs ‘multiple trigger wheeze’);17 and atopic status (namely ‘atopic’ vs ‘non-atopic’).18 Paediatricians should be honest to admit that they have been naive a long time in thinking that atopy is ‘all-or-none’ concept. Indeed, it is now clear that IgE antibody responses do not reflect a single phenotype of atopy, but several different atopic vulnerabilities that differ in their relation with asthma presence19 and the risk of a loss of lung function throughout childhood.20
Although these phenotype classifications are attractive since they are based on patient characteristics at baseline, they have not been validated prospectively to identify individuals responding to specific therapeutic approaches and they are generally not useful in guiding treatment. An additional, intrinsic limitation is the substantial between physician variability in phenotype assessment.21 Yet, in the European Respiratory Society (ERS) classification children may not fit clearly into either category and there is over time overlap between the two phenotypes that represent the extremes of a clinical spectrum and reflect the multifactorial nature of wheeze.22 The ERS classification establishes a framework for stratifying children in therapeutic trials; however, it begs the question whether the multiple trigger phenotype is the same condition as childhood asthma.23 The answer depends on what definition of asthma is being used. If the definition includes the presence of airway eosinophilic inflammation, the evidence is that airway pathology in children with multiple trigger wheeze, at least in severe cases, is similar to childhood and adult asthma.24 Indeed, the ability to predict in a young child the persistence of wheeze up to school age has been largely evaluated in the literature.25 An ideal, predictive method does not exist yet, but a simple, practical and non-invasive tool for predicting asthma at school age in young children with wheeze or cough has been proposed recently.26 Unfortunately, there is accumulating evidence that the disappearance of symptoms does not necessarily correlate with the absence of respiratory sequelae.27
On the edge of reason
The correct treatment of a disease should be preceded as far as possible by a correct diagnosis. Scientific medicine (for instance, evidence-based therapy) is very largely dependent on placing patients as closely as possible into disease pigeonholes. Clearly, neither one extreme—every wheezy child has a unique phenotype—nor the other—placing all wheezy children in the same disease entity ‘asthma’—provides much guidance for prevention, prognosis or treatment. A utilitarian position falls between these two extremes.28 Unfortunately, despite the report's care in outlining the limitations of the ERS classification, physicians maintained a faithful—and sometimes uncritical—behaviour in prescribing treatments for this common disorder; a rather one-size-fits-all mindset, wheeze mandates a prescription for inhaled corticosteroids (ICSs).
In a study that examined dispensing and co-dispensing of antibiotics and asthma drugs in 892 841 Belgian children, an asthma medication was dispensed to 45% of children aged <3 years, 18% aged 3–7 years and 8% aged 8–18 years. In more than one-third of the children, the medication(s) delivered included ICS, irrespective of the age range. Surprisingly, chronic use of ICS (ie, >2 canisters) was almost three times (28% vs 11%) more frequent in young children than in school-age children.29 Clearly, ICS prescription was highest in an age category for which the least evidence for asthma drug efficacy has been documented.30 In a recent Dutch study, general practitioners' prescribing behaviour to children with respiratory symptoms was also characterised by inappropriate prescription of ICS.31 Mainly, this attitude was driven by doctors' beliefs about ICS that differed from currently available evidence. This paradoxical prescription of asthma medications in young children is not an isolated phenomenon. Indeed, in preschool children across Europe and the USA, wheezing with cough and/or breathlessness, which is known to be precipitated mainly by viral infections, is treated with antibiotics in 34% of cases.32 The finding that antibiotics are prescribed twice as frequently to children who receive asthma medications29 ,32 is another aspect that is difficult to justify or understand, and that must in part be attributable to inappropriate practice.
The Manichean doctrine of good and evil is hard to apply to clinical medicine. Indeed, in a birth cohort study in young children, acute wheezing episodes were significantly associated with bacterial infections with similar frequency to, but independent of, the association with viral infections.33 Yet, it was recently shown that detection of specific bacterial pathogens such as Moraxella catarrhalis and Streptococcus pneumoniae was more common than viruses in children with respiratory illnesses, and asthma exacerbations resulted to be more severe when coinfection with rhinovirus was present.34 These studies do not definitively prove the causative role of bacteria, and only controlled, clinical trials can elucidate the therapeutic role of antibiotics in wheezing illness. Should this role—apparently heretical so far—be evident, it may result in an step forward for treating a condition characterised by disappointing results from different antiasthma therapies. In a recent double-blind, placebo-controlled trial in children aged 1–3 years with asthma-like symptoms, a 3-day course of oral azithromycin reduced the duration of episodes, suggesting that this drug could have a role in acute management of exacerbations.35 Further research is needed to disentangle the inflammatory versus antimicrobial aspects of this relationship. However, in view of the finding of bacteria in the majority of wheezy episodes, the common term ‘viral wheeze’ seems inappropriate.
Infantile wheeze is responsible for a disproportionately high consumption of economic resources worldwide. When we deal with a wheezy infant, how much are we in fact treating the parents who ‘want something done’ about the respiratory sounds made by their child? It is well known that illness perception influences the way in which patients with asthma cope and self-manage of the illness.36 And again, what can we do for wheezy children in terms of improving their condition? First, it is important to consider whether the child really needs the treatment. Unfortunately, other respiratory sounds are frequently mislabelled by parents as wheeze,37 and this obviously increases the risk of unnecessary prescription of medication in clinical practice. Confirming parental reports of wheeze by an experienced doctor38 should not be considered a purely intellectual exercise, but is instead the watershed between a rational and a careless medical act.
Despite consistent scientific support on the association of exposure to tobacco smoke39 and air pollution40 with triggering of wheezing symptoms in the first years of life, attention to environmental measures continues to be largely overlooked by parents.41 As physicians, we should never tire of repeating that we ought to spend much more of our time in discussing environmental prevention actions with the parents of a wheezy child before writing down any medication in our prescription book.
The approach to treatment of intermittent wheeze among preschoolers has been traditionally based on that of acute asthma in older children. The scenario—and the response to treatment accordingly—is however complicated by the heterogeneity of early childhood wheeze/asthma, potentially related to different patterns of airway inflammation. Physicians should be aware of this in their clinical practice. In principle, intermittent symptoms should be treated with intermittent therapy. Whatever the nature of wheeze, and although the evidence supporting the usefulness of these drugs in young children is limited,42 intermittent use of bronchodilators, either β2-agonists or anticholinergics delivered by a metered-dose inhaler and a spacer, is the treatment of choice for acute episodes. There is no role for nebulised therapy to deliver bronchodilator apart from in children too sick to use inhalers. A short trial of bronchodilator therapy is indicated for happy wheezers, but therapy should be stopped in the absence of clinical improvement. Due to their established efficacy in acute asthma,43 oral corticosteroids have been used for decades to treat episodes of wheeze in young children. This practice has been recently questioned. Two rigorous studies, conducted at home44 or in hospital,45 have shown no benefit of oral prednisolone over placebo in preschool children. It should be noted that many had relatively mild wheeze, with 75% being discharged from hospital within 24 hours,45 and that the dose of steroids may have been too low with respect to other studies.46 ,47 However, the evidence is not strong enough to determine whether lower-dose regimens are generally less effective than higher-dose regimens.48 The persistent controversy of the efficacy of oral corticosteroids in the treatment of preschool wheeze has not been completely resolved by a recent meta-analysis,49 but it is difficult to argue that oral corticosteroids have not in the past been overprescribed in this clinical setting. There may be a steroid-responsive subgroup, for example, particular viral infections, and this needs to be determined. Oral corticosteroids must not be given for all attacks of viral wheeze, but a short burst should be considered in children admitted to hospital with features suggestive of a severe attack of atopic asthma (eg, a combination of multitrigger wheeze, severe eczema and a family history of atopic asthma; note that atopy was not an exclusion criterion in the hospital trial45) or with really severe bronchodilator-unresponsive wheeze where high-dependence care is needed.
If there is a failure to control symptoms with bronchodilator therapy, intermittent therapy with high-dose ICS50 ,51 starting at the first sign of respiratory tract infection and continued for up to 7–10 days might be a useful approach. Our daily experience suggests that many wheezy children are treated with pre-emptive ICS, but at lower doses than have been proven to be effective.52 In practice, we suggest that the fluticasone dose should not exceed the licensed dose of 150 μg twice a day by a metered-dose inhaler and a spacer, given the number and duration of viral colds in normal preschool children and the risk of side effects including growth suppression and adrenal failure with higher doses. The potential for overuse of intermittent high-dose ICS needs to be evaluated properly. The effectiveness of intermittent montelukast for wheeze in young children is unclear50 ,53 and is likely limited to a specific genotype-dependent subgroup.54 There are currently no studies that have combined intermittent high-dose ICS with intermittent montelukast to treat episodic viral wheeze.
There should be strict indications to prescribe a controller therapy in a young child with recurrent wheezing. The recent ERS Consensus3 and Global Initiative for Asthma guidelines55 agreed that the frequency of symptoms (on most days of the week, responding to β2-agonists) and their severity (ie, prolonged disruptive symptoms that required repeated emergency visits or admission to hospital) are the two main reasons for a trial of any controller therapy (ICS or montelukast), irrespective of the phenotype. The scientific evidence in this setting is scanty, in particular to support the use of regular ICS in preschool children with episodic viral wheeze;56 is the siren cry to do this an example of physicians wanting ‘to do something’ even if it is ineffectual, rather than lose face by admitting therapeutic bankruptcy.
Whatever the clinical context and the chosen medication, there are fixed cardinal points. First, the fact that regular treatment with ICS, even at low dose, may lead to growth suppression in young children57 should be discussed with parents. Second, a therapeutic trial should always be given for a fixed time period (such as 6–8 weeks) and discontinued at the end of the agreed period to see whether symptoms recur or have resolved, and treatment has become unnecessary. Third, the children must be reviewed regularly to evaluate the response to treatment and any changes in symptom pattern. Fourth, education of the parents in the correct use of inhalers—a practice that is often neglected58—should be repeated at each visit.59 If an inhaled drug does not seem to be working, checking that it is being properly administered rather than escalating treatment is mandatory.
Finally, physicians (and families too) should be aware that no drug strategy is available to reduce future risk of asthma, and that failure to initiate regular treatment will not prejudice future respiratory health of the child. Indeed, neither early use of continuous60 ,61 nor intermittent with viral colds62 ICS can prevent the progression of any phenotype of preschool wheeze to asthma in childhood.
Can we do better?
We have unfortunately gone in circle in the past; Martin Luther, who memorably is said to have stated, “Human reason is like a drunken man on horseback; set it up on one side, and it tumbles over on the other”, would have recognised this. How do we break this circle? Rather than endless discussion and using words loosely, let us describe what we can see (or measure) in the preschool wheezer. The relevant components of preschool airway disease are fixed and variable airflow obstruction, inflammation and infection.63 Spirometry can be performed in two-thirds of 2–5-year-old childen.64 Reversibility testing is harder to do in this age group since the changes with bronchodilators may be less than the baseline variance of the method,65 and lung clearance index is not useful in this context.66 Induced sputum can be obtained even in very young children and gives a better indication of any bacterial infection;67 however, it poorly reflects bronchoalveolar lavage eosinophilia in this age group, unlike in older children.68 Peripheral blood eosinophils do seem to provide an accurate reflection of lower-airway eosinophilia in this age group.67 They may be the indicator of other atopic disorders such as eczema or rhinitis, but in the absence of peripheral blood eosinophilia, prescribing ICS to a preschool child who wheezes is unjustified. Unfortunately, the majority of the diagnostic tests are impractical for routine clinical use, although bigger studies are needed to assess the utility of the readily available blood eosinophil count. Hopefully better biomarkers will be available in the near future, but until then surely a blood eosinophil count is within the grasp of every treating paediatrician and primary care doctor. How ridiculous in the 21st-century endlessly to discuss words and use words-based treatment when data-driven therapy should be available.69
Lessons for the future
Looking to the past, over the last half century there have been no words from the Messiah about the best way of managing wheezy children. However, using the understanding gained by great thinkers who have gone before has certainly allowed us to see things that are more distant.
The long story of infantile wheeze taught us that to be confident about a black and white system in medicine is an intellectually poor position, and that indeed we should be ready to modify our perspectives according with the evolution of the scientific progress. Medicine advances by challenging dogmas rather than uncritically accepting them. Wisdom is also to shift thinking as evidence builds. Finally, scientific progress should not allow us to forget that treating the patient—in this case, a child—and not a disease remains the main principle of the art of the medicine.
The authors thank Mrs Consuelo Ramacogi for the administrative, technical, and material support.
Contributors FMdB wrote the initial draft of the manuscript. Both authors participated in the critical revision for intellectual content and accepted the final version of the manuscript.
Funding AB was supported by the NIHR Respiratory Disease Biomedical Research Unit at the Royal Brompton and Harefield NHS Foundation Trust and Imperial College, London.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.