Childhood asthma and infection: virus-induced exacerbations as determinants and modifiers

Although there is increased awareness of the importance of information on the role of infections in the initiation, clinical presentation and possibly persistence of asthma and related reactive airway syndromes, systemisation of this knowledge in a way helpful for disease management remains suboptimal. Several conceptual issues have to be taken into account: the fact that the effects of an infection may vary according to genetic background, the current immune status of the host, and parallel environmental stimuli, in addition to the particular infectious agent itself. Interestingly, these effects can be nonlinear, and even showing completely opposite results at different exposure levels 1. Different aspects of the disease, such as control, severity, exacerbation and persistence, may not be pathophysiologically identical and, therefore, each outcome should be addressed individually. Furthermore, the nature and limits of asthma are disputed, often at a philosophical level (is there true asthma, in the Platonic sense?) Childhood is a particularly challenging period, due to the fast and continuous maturation processes involving, among others, the immune, endocrine, neural and respiratory systems. From a clinical perspective, variable patterns or phenotypes may be the result of different underlying mechanisms; however, these remain to be characterised in detail. In the present article, the current evidence is reviewed regarding the associations between infection and exacerbations of established asthma, an area with a relative abundance of data and somewhat better understanding. Effects of infection on disease initiation will be presented in a subsequent article in the present series.SERIES “HOT TOPICS IN PAEDIATRIC ASTHMA”Edited by K-H. Carlsen, G. Hedlin and A. BushNumber 2 in this Series

DEFINITIONS

Although a considerable proportion of asthma morbidity, mortality and health costs can be attributed to exacerbations, objective criteria for defining such events are lacking, and it is sometimes difficult to differentiate an exacerbation from poor asthma control 2–5. The lack of a universally accepted definition of an exacerbation creates difficulties in the interpretation of clinical research and trial results, with the exception of severe asthma, exacerbations being defined fairly obviously 6.

The Global Initiative for Asthma guidelines define exacerbations as episodes of progressive shortness of breath, cough, wheezing or chest tightness, or a combination of these symptoms 7. Exacerbations have also been categorised as mild, moderate, severe or life-threatening. Variability in peak flow or forced expiratory volume in 1 s and other lung measurements have been suggested to define exacerbations, although with poor precision 8. Several scoring systems, with respect to severity, combining a number of physical signs and symptoms have often proven inaccurate when used in children 9. A working definition of an exacerbation that has been applied in several trials of pharmacological therapies has been a worsening of asthma of sufficient severity to require intervention from a medical professional or self-administration of oral corticosteroids 10. A joint task force of the American Thoracic Society and European Respiratory Society has recently defined asthma exacerbations as events characterised by a change from the patient’s previous status 4. Severe exacerbations were defined as events that require urgent action to prevent serious outcomes, such as hospitalisation or death, whereas moderate exacerbations as events troublesome to the patient that prompt a change in treatment and are outside the patient’s usual range of day-to-day asthma variation; however, such an exacerbation is difficult to differentiate from poor asthma control requiring additional treatment, as stated previously. Mild exacerbations were not defined because such events can be indistinguishable from loss of asthma control 4. From a paediatric perspective, these definitions are more difficult to use, taking into account the dependence on parental reporting, as well as the variability among childhood age groups in both pathophysiology and medical practice.

Although several studies have shown that asthma exacerbations are strongly associated with respiratory tract infections, and the term virus-induced exacerbation is not uncommon, only a small number of such studies were prospective 11, 12, and even fewer have simultaneously assessed other potential factors that may contribute to an exacerbation 13. Furthermore, respiratory infections do not always result in an exacerbation, and there is little evidence that treating or preventing the infection may cure or prevent an exacerbation. Therefore, a causal relationship in the sense of fulfilment of Koch’s postulates is difficult to establish. However, in the PRACTicing ALLergology (PRACTALL) consensus report, evident post-infectious asthma activity was used to phenotype children under the term virus-induced asthma 14. More recently, a European Respiratory Society task force defined episodic viral-induced wheeze in preschool children as a distinct phenotype of asthma-related symptoms 15. Whether or not repeated episodes of post-infectious wheeze in very young children should be considered asthma remains the subject of debate. However, the same pattern frequently appears in older children with diagnosed asthma, either as predominantly virus-associated episodes or within a multi-trigger syndrome, strongly associated with atopy 16.

The obvious need for delineation of different patterns, or phenotypes, to improve definitions, and consequently clinical practice, is particularly hampered in childhood due to the considerable overlap between patterns and the rapidly evolving natural history. Nevertheless, the clinical experience of acute asthma worsening preceded by a common cold is overwhelming in paediatrics, and confirmed by many epidemiological studies mentioned below. Therefore, although the limits of virus-associated asthma exacerbations in children will continue to be scrutinised and debated, the majority of such events would be clinically indisputable.

EPIDEMIOLOGY

The development of sensitive diagnostic PCR-based techniques permitted demonstration of an already clinically suspected association between common viral respiratory infection and asthma exacerbations. Respiratory viruses have been identified in 80–85% of exacerbations resulting in reduced peak expiratory flow and wheezing in school-aged children, and 45–80% of adults, with rhinoviruses (RVs) being the most frequently detected (60% of cases of asthma exacerbation) 11, 17, 18. More recent studies have suggested that RV, beside being the most prevalent, is also the only virus significantly associated with exacerbations in children (odds ratio 6.8) 19. Interestingly, the peak of severe exacerbations in children occurs shortly after their return to school following breaks, and 1 week later in adults, and this coincides with peaks in RV infections in the autumn and, to a lesser extent, spring 20. In the aforementioned study, the prevalence of virus identification, predominantly RV, was 61% in subjects presenting with an asthma exacerbation compared to 40% in asthmatic controls. There are suggestions that concomitant exposure to allergens (e.g. animal dander) at school could contribute to loss of asthma control 21. It should be noted, however, that the proportion of hospitalisations for asthma requiring admission to an intensive care unit remains constant relative to total admissions 22. Other respiratory viruses responsible for exacerbations display different seasonal variation; respiratory syncytial virus (RSV) and influenza have been associated with increased asthma exacerbations during the winter months 23.

INFECTION AMONG FACTORS PRECIPITATING ASTHMA EXACERBATIONS

Environmental factors, including pollutants, allergen exposure, psychological stress and weather conditions, may increase the risk of developing an acute asthma exacerbation in both children and adults 24–30. The multiplicity of precipitating factors points towards a multifactorial cause, as well as potential interactions, which have been shown for some combinations of factors, such as pollutants 31, 32. The risk of hospitalisation due to asthma exacerbation is synergistically increased in the presence of viral infection, allergen sensitisation and exposure to respiratory allergens 33. Moreover, studies using sophisticated statistical methods have emphasised the importance of psychological events in asthma control, showing that stressful life events have immediate and delayed effects in significantly increasing the risk of an asthma exacerbation 34. Nevertheless, no studies have been able to assess all candidate-precipitating factors simultaneously in order to evaluate their relative importance and interactions. The classical notion that asthma exacerbations in atopic patients are solely related to allergen exposure with subsequent inflammation has been challenged 35.

Respiratory viral infections are the factor most strongly associated with precipitation of acute asthma exacerbations in various populations 36–38. RVs are the predominant aetiological agents associated with exacerbations; however, other viruses, such as coronavirus, influenza virus, adenovirus, RSV and parainfluenza virus, Mycoplasma pneumoniae and Chlamydophila pneumoniae have been also identified, although to a lesser extent 39. Among the more recently identified viruses, human metapneumovirus has been associated with wheezing episodes, mostly in younger children, whereas bocavirus claims ∼5% of asthma exacerbations 40–42. Nevertheless, RV remains by far the major player in asthma exacerbations; it has also been associated with bronchiolitis, although clearly distinct from asthma-related diseases, and pneumonia 43, 44. Recent studies have suggested that RV wheezing illnesses in infancy and early childhood are among the major predictors of the subsequent development of asthma at the age of 3 and 6 yrs in high-risk children 45, 46. RV serotypes were traditionally classified into two groups based on viral protein (VP) 1 and VP4/VP2 sequences, RV-A comprising 74 different serotypes and RV-B 25 serotypes. More recently, using PCR nucleotide sequence homology and phylogenetic analysis, a third group of RVs, RV-C, has been described, including at least nine serotypes 47. Febrile wheeze and asthma exacerbation were the commonest presentation (76%) of RV-C infection, whereas others have shown strong associations of this novel RV with bronchitis, bronchiolitis and pneumonia 48, 49. In a prospective study of preschool children hospitalised with acute respiratory illness or fever, RV-C infections accounted for 46% of RV-positive samples and were significantly associated with an asthma diagnosis at hospital discharge compared to patients with RV-A (55 versus 36%, respectively) 50. This result has now been duplicated in various settings, suggesting that this novel group might cause a substantial burden in paediatric asthma 51–53.

M. pneumoniae and C. pneumoniae are found more frequently in the airways of patients with asthma than in healthy controls; their role in exacerbations is less clear 54. In the majority of exacerbation studies looking for a wide array of pathogens, the relative proportion of atypical bacteria is rather low. However, M. pneumoniae was identified in 20% of exacerbations in asthmatic children requiring hospitalisation, and in 50% of children experiencing their first asthmatic attack 55. More recently, atypical infection has been associated with more severe asthma exacerbations, as shown by spirometry and symptom severity scores 56. In support of the hypothesis that asthmatic airways are more susceptible to M. pneumoniae infection, Toll-like receptor 2 expression and interleukin (IL)-6 production are downregulated by M. pneumoniae infection during ongoing ovalbumin-induced allergic inflammation 57. Moreover, airway epithelial cells from asthmatic subjects show increased production of mucin 5 subtypes A and C (the major airway mucin) when infected with M. pneumoniae compared to normal controls 58. These data suggest that pre-existing inflammation might lead to decreased clearance of M. pneumoniae, thus increasing airway inflammation. However, whether or not this may result in symptoms, exacerbations or simply airway colonisation remains to be elucidated.

MECHANISMS OF VIRUS-INDUCED EXACERBATION

A wide range of mechanisms has been implicated in the progression from a viral respiratory infection to an acute exacerbation of asthma. These have recently been reviewed in detail 59. The major and minor group of RVs bind to intercellular adhesion molecule (ICAM)-1 and very low-density lipoprotein receptors on the respiratory epithelium, respectively, which induces viral internalisation and upregulation of additional receptors 60. The bronchial epithelium is the site of viral replication 61 and participates in the initiation of antiviral responses 62, 63, innate and adaptive immune responses and inflammation (fig. 1) 64. RVs are also able to pass into the bloodstream, causing viraemia, more frequently during acute and severe asthma exacerbations 65.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1–

Overview of the mechanisms involved in rhinovirus (RV)-induced asthma exacerbations. A variety of mediators produced by epithelial and immune cells following a viral infection result in an inflammatory response, including chemotaxis of neutrophils and eosinophils. An antiviral response is also mounted by both epithelial and dendritic cells, producing interferons (IFNs). In parallel, a number of epithelial-derived growth factors may contribute to airway remodelling. ICAM: intercellular adhesion molecule; IL: interleukin; TNF-α: tumour necrosis factor-α; GM-CSF: granulocyte-macrophage colony-stimulating factor; CCL: CC chemokine ligand; CCL5: RANTES; CXCL: CXC chemokine ligand; CL11: eotaxin; CCL24: eotaxin-2; CXCL10: IFN-γ-inducible protein 10; TLR: Toll-like receptor; pDC: plasmacytoid dendritic cell; mDC: myeloid dendritic cell; FGF: fibroblast growth factor; VEGF: vascular endothelial growth factor.

RV infection activates inflammatory pathways by increasing levels of neutrophils, eosinophils, CD4+ cells, CD8+ cells and mast cells through increased mRNA expression and translation of IL-6, IL-8, IL-16, eotaxin, interferon (IFN)-γ-inducible protein (IP) 10, CC chemokine ligand 5 (RANTES) and other pro-inflammatory cytokines and chemokines 66. Moreover, viruses generate systemic as well as neural responses, further inducing inflammation and hyperresponsiveness.

Virus-induced asthma exacerbations are mainly characterised by neutrophilic inflammation 67. Evidence of neutrophilic degranulation and increased lactate dehydrogenase levels are independent predictors of severity, and increased levels of the potent neutrophil chemokine IL-8 are found in exacerbations 68, 69. It has been suggested that neutrophil activation and degranulation are associated with clinical severity in virus-induced asthma, probably via neutrophil protease-induced mucus secretion by airway gland serous cells 68, 70. RV infection also results in the release of IP-10, a chemokine involved in T-cell recruitment and mast cell activation. Increased serum IP-10 levels in atopic asthmatics were strongly associated with severe airway obstruction and a reduced bronchodilator response to β₂-agonists. Finally, increased serum IP-10 levels, in combination with low TNF-α levels, were predictive of a virus-induced cause as the exacerbation trigger 71.

Until recently, a major obstacle to research into the mechanisms of RV infection has been the lack of an animal model, since major RVs would not bind mouse ICAM-1 receptors 72. In 2008, three novel mouse models for both major and minor serotypes of RV, presenting features similar to those observed in RV infection in humans, were developed 73. Transgenic BALB/c mice with a mouse/human ICAM-1 chimera were successfully infected with the major group virus RV16. These mouse models are characterised by several relevant asthma-related outcomes, such as airway neutrophilic and lymphocytic inflammation, mucus secretion and the induction of mediators also induced by human RVs. Moreover, RV1B, a minor receptor subtype, was capable of binding the mouse receptor and inducing successful infection in vivo; increased airway inflammation and hyperresponsiveness has also been documented. Phosphatidylinositol 3-kinase/Akt kinase signalling was required for maximal RV1B-induced neutrophilic inflammation, probably via its essential role in virus internalisation 74.

Innate immune responses, and in particular IFNs, seem to play a major role in RV-induced asthma exacerbations. Type-I IFNs (α/β) are important components of the innate immune response, inducing an antiviral state in infected and nearby cells, via induction of apoptosis and proteins with direct antiviral activity 75, 76. More recently, a novel class of IFNs has been discovered, named type-III IFNs, including IFN-λ1 and IFN-λ2/3 77, with a potentially immunoprotective role in the lower airways 78.

In vitro studies have shown that asthmatic bronchial cells are profoundly deficient at producing RV-induced IFN-β, whereas exogenous IFN-β induced apoptosis of RV-infected asthmatic bronchial epithelial cells (BECs) and restoration of innate immunity by inhibition of RV replication, as observed in normal BECs 79.

It has been suggested that defective innate immune responses result in deficient adaptive type-1 T-helper cell (Th) responses. In support of this, peripheral blood mononuclear cells (PBMCs) from asthmatics have been shown to exhibit a deficient IFN-γ response to RV; moreover, RV-stimulated PBMCs from atopic asthmatics resulted in reduced production of IFN-γ and increased IL-4 and IL-10 compared to normal controls 79, 80. An inverse correlation between IFN-γ production and viral load and/or symptoms score has been documented 81. In the same context, PBMCs from asthmatic patients were reported to produce less IFN-α2 than PBMCs from normal controls in response to RSV and Newcastle virus infection, also implicating IFN-α in the pathogenesis of asthma exacerbations 82.

With respect to type-III IFNs, it has been documented that RV induces IFN-λ production in a suboptimal way, in both primary BECs and macrophages from atopic asthmatic individuals. The deficient type-III IFN production strongly correlated with viral replication in vitro. Subsequent experimental infection with RV-16 resulted in deficient induction of IFN-λ in bronchoalveolar lavage cells from asthmatics; such induction was strongly inversely associated with severity of clinical illness, reductions in lung function and increased airway hyperresponsiveness 83.

Both type-I and type-III IFNs are induced by RV and influenza infection in BECs. IFN-λ appears to be the most sustained and abundant IFN subtype induced in the lung by RV infections, whereas, in PBMCs, all IFNs were induced to a similar degree. It has been suggested that IFN-λ could serve as a candidate therapy for asthma exacerbations, and possibly other viral diseases in which BECs are the major site of virus replication 84.

Despite the importance of viral infections in asthma exacerbation pathogenesis, a combination of interacting factors is likely to be involved. Recent studies have emphasised the importance of atopy and allergic inflammation in the induction and perpetuation of virus-induced respiratory diseases 85, 86. Allergens and respiratory viruses act synergistically in the expression of asthma symptoms in adults and children 13, 33. RV-induced epithelial cytotoxicity could increase the penetration and effects of allergens, leading to further induction of inflammation 87, 88. Synergy between virus and mite (Dermatophagoides pteronyssinus antigen 1)-induced inflammation may also occur through combined nuclear factor-κB activation 89. In primary BECs, RV infection results in efficient viral replication, leading to cell lysis, which is greater in cells from atopic asthmatics than from healthy controls. Such a response could be attributed to the impaired induction of early apoptosis found in the asthmatic cells, probably due to reduced levels of IFN-β 79. In line with these observations, it has been shown that, under the influence of an atopic immune response, the epithelial inflammatory response to RV becomes suboptimal with respect to cytokines and chemokines, such as IL-6, IL-8 and RANTES. Such a response is associated with inadequate viral clearance and increased RV-induced cytotoxicity 90. However, the timing of the infection might play a significant role, since simultaneous infection and allergen exposure significantly increase the severity of asthma exacerbations in asthmatics, whereas allergen exposure prior to infection has minimal or no effect on symptoms or airway hyperresponsivess 33, 91.

In allergen-sensitised and -challenged mice, RV infection exacerbates neutrophilic, eosinophilic and lymphocytic airway inflammation and hyperresponsiveness, mucus secretion and production of both Th1 and Th2 cytokines. Such responses are consistent with the notion that allergic inflammation substantially exacerbates RV-induced inflammation 73.

From another perspective, RV infection has been shown to promote endothelial cell proliferation and differentiation though induction of angiogenetic factors, such as vascular endothelial growth factor (VEGF), possibly contributing to airway remodelling in asthma; this response was further augmented in an atopic host environment 92. Similar responses were observed for other profibrotic factors, such as transforming growth factor-β 90 and fibroblast growth factor 2 93. These data indicate a mechanism through which viral respiratory infections may promote airway remodelling and the chronic asthma phenotype in atopic asthmatic individuals.

TREATMENT

Despite the impact of virus-induced asthma exacerbations, no specific antiviral treatment able to modify the clinical outcome of the viral infection exists. Pharmacological interventions shown to be relatively successful at preventing asthma exacerbations include inhaled corticosteroids and their combination with long lasting β₂-agonists (mostly in adults), leukotriene modifiers and anti-immunoglobulin E therapy 8, 94–97. Corticosteroids suppress eosinophilic inflammation through various pathways 98, 99. However, the inability of corticosteroids to modulate several other aspects of the immune response to respiratory viruses that lead to asthma exacerbations make them not as effective at preventing or treating virus-induced asthma exacerbations in different age groups 100. Leukotriene receptor antagonists, used either prophylactically or therapeutically, exhibit a modest effect on virus-induced asthma exacerbations in children 94, 101, 102. In conclusion, treating virus-induced asthma exacerbations is far from satisfactory, despite current knowledge regarding the underlying pathogenesis 103.

The antigenic diversity of RVs makes the development of a vaccine rather unlikely. Modulation of the host response to virus infection is a potential approach to therapy; taking into account the impaired IFN production in asthmatics, IFN supplementation could be a targeted treatment for asthma exacerbations 6. Experimental data from in vitro models have shown that the administration of exogenous IFN-β induced apoptosis and suppressed viral replication, suggesting the potential for therapeutic intervention 79. Attempts with intranasal IFNs (IFN-α2 and IFN-β-serine) have not found their way to the clinic to date, due to either reduced efficacy or severe side-effects 104, 105. Pleconaril is an agent with broad-spectrum activity against RVs, which has been developed as an oral preparation to be taken twice daily. Although modest efficacy against natural colds has been verified in phase-II clinical trials, concerns have been raised about the development of resistant RV mutants 106, 107. At this stage, there are no clinical data regarding the efficacy of pleconaril in acute asthma exacerbations.

Macrolides and ketolides, beside their antimicrobial action, possess important immune-modulating effects 108–110. A multicentric double-blind randomised placebo-controlled study assessed the efficacy of oral telithromycin for 10 days, as a supplement to standard treatment for adults with an acute asthma exacerbation. Significant improvements in both symptoms and lung function were shown in the treatment group, and the effect did not seem to depend crucially upon the presence of atypical bacteria 111. Further work-up is essential in order to identify patient groups likely to benefit from similar treatment approaches among children 112. Such a study of clarithromycin in acute exacerbations of asthma in school-age children is underway (data not shown). With respect to RSV-induced lower respiratory tract diseases, recent data indicate lack of efficacy of azithromycin in hospitalised children aged <24 months 113.

A variety of novel options for the prevention and treatment of virus-induced asthma, such as antioxidants, VEGF and angiogenesis inhibitors, and more, need to be further assessed for their efficacy and applicability.

CONCLUSIONS

Viral infections are the major precipitant of asthma exacerbations, leading to a complex of inflammatory processes. Although knowledge of the mechanism underlying infection-induced asthma exacerbations has increased substantially since the late 1990s, a great deal of further work is still clearly warranted. Moreover, the interactions between viruses, other pathogens, air pollution, and allergen sensitisation and exposure are incompletely understood. Identification of the mechanisms involved in the pathogenesis of asthma exacerbations should facilitate development of future treatments tailored to the underlying cause.