|Home | About | Journals | Submit | Contact Us | Français|
Associations between respiratory infections and asthma inception and exacerbations are well established. Infant respiratory syncytial virus and rhinovirus infections are known to be associated with an increased risk of asthma development, and among children with prevalent asthma, 85% of asthma exacerbations are associated with viral infections. However, the exact nature of this relationship remains unclear. Is the increase in severity of infections an epiphenomenon, meaning respiratory infections just appear more severe in individuals with underlying respiratory disease, or instead a reflection of altered host susceptibility among persons with asthma and atopic disease? The main focus of this review is to summarize the available levels of evidence supporting or refuting the notion that persons with asthma or atopic disease have an altered susceptibility to selected pathogens, as well as discussing the biological mechanism(s) that might explain such associations. Finally, we will outline areas in need of further research, as understanding the relationships between infections and asthma has important implications for both asthma prevention and treatment, including potential new pathways that might target host immune response to select pathogens.
A role for respiratory infections in asthma development and asthma exacerbations has been well documented (Fig 1). Infections in infancy, including respiratory syncytial virus (RSV) and rhinovirus (RV), are known to be associated with an increased risk of asthma development(1). Furthermore, this has been demonstrated to be a dose response relationship, with severe episodes of infant bronchiolitis increasing the odds of both early childhood asthma and asthma-specific morbidity(2). The same relationship exists between viral respiratory tract infections and asthma exacerbations(3;4).
Asthmatics have both increased frequency and severity of lower respiratory tract infections as compared to individuals without asthma (Fig 1). In a prospective study of RV infections in cohabitating couples consisting of one asthmatic and one non-asthmatic subject those with asthma had more frequent lower respiratory tract infections, along with symptoms which were more severe and of greater duration(5). In contrast, asthmatics(5) and those with atopic disease(6) do not appear to have increased frequency, severity, or duration of upper respiratory tract infections.
Asthma has also been identified as a risk factor for influenza-attributable morbidity and community acquired pneumonia. Influenza-attributable health care utilization, including both outpatient visits(7) and hospitalization rates(8), is higher among asthmatic children as compared to healthy children, as are influenza-related complications(9). Population-based surveillance for laboratory-confirmed influenza hospitalizations, found children with asthma to account for 32% of influenza-associated hospitalizations during the 2003-2009 influenza seasons and 44% during the 2009 H1N1 pandemic, four to five times the asthma prevalence rate(8). During the 2009 H1N1 pandemic asthmatics, as compared to non-asthmatics, were almost twice as likely to have pneumonia (50% vs. 27%) and require care in the ICU (33% vs. 19%)(9). However, in a study conducted in the UK, while asthmatics were more likely to have severe respiratory distress and require supplemental oxygen, they were half as likely to die or require an advanced level of care as compared to non-asthmatics(10). The authors found that the less severe outcomes in asthmatics were associated with prior inhaled steroid use and earlier hospital admission. Thus, asthmatics still may be at an increased risk for influenza related complications, but due to their medical history more likely to seek care earlier, resulting in improved outcomes. The risk for community acquired bacterial and viral pneumonia has been estimated to be at least two fold in patients with asthma as compared to healthy controls(11-13).
Asthmatics with allergic sensitization appear to have an even greater susceptibility to respiratory infections. Children with atopic asthma were found to experience 47% more symptomatic viral illnesses as compared to non-atopic asthmatics during peak virus season (1.19 vs. 0.81 per month)(14). Allergen sensitized asthmatics also have a higher risk of hospital admission for asthma exacerbations as compared to non-sensitized asthmatics(15).
The increased susceptibility to infections among persons with asthma extends beyond the lungs (Fig 1). Children with asthma have increased rates of both otitis media and gastroenteritis during infancy. Increased prevalence of ear infections during infancy was reported by Eldeirawi et al(16) among Mexican American asthmatic children (39%) as compared to non-asthmatic children (20%); this association was independent of antibiotic use and other infectious history. A similar relationship was noted between otitis media and both asthma and atopic dermatitis in a German birth cohort(17). A cross-sectional study of 26,400 Korean children by Ahn et al(18) identified a higher prevalence of acute gastroenteritis during infancy in children later diagnosed with asthma. Recall bias is a concern for the studies conducted by Eldeirawi(16) and Ahn(18) as they relied on retrospective reporting. In addition, patients with atopic dermatitis are highly susceptible to cutaneous bacterial, viral, and fungal infections, most notably S. aureus and herpes simplex virus(19;20). Delayed therapeutic response times and an increased likelihood of recurrence were identified for genital warts in patients with a history of hay fever, eczema, or asthma(21).
Bacterial colonization of the airways in infancy is associated with asthma development (Fig 1). Bisgaard et al(22) collected hypopharyngeal samples from 321 asymptomatic neonates at 1 month of age and found colonization of the airways with S. pneumoniae, H. influenzae, or M. catarrhalis to be associated with the development of asthma by age 5 years. We postulate this colonization is due to an altered immune response that predisposes these infants to early acquisition of these pathogens, although this colonization could also indicate a causal relationship between the pathogen and asthma inception. Increased frequency of colonization with S. pneumoniae has also been demonstrated in asthmatics. In a cross-sectional, population-based prospective study of 1,013 adolescents, asthma was an independent risk factor for nasopharyngeal colonization of S. pneumoniae, identified in 8.2% of subjects(23). An increased prevalence of bacterial colonization of the skin, primarily with S. aureus, is seen in patients with atopic dermatitis(24) and occurs on both lesional and clinically normal skin(25).
A number of latent infections have been demonstrated to be more common among subjects with asthma, including M. pneumoniae(26;27), C. pneumoniae(27;28), adenovirus(29;30), and RV(30-33). We have defined latent infections as asymptomatic bacterial or viral identification after an acute initial infection. As with colonization, we propose latent infections reflect an altered immune response, although an alternative explanation is that these pathogens may play a role in asthma pathogenesis. Mycoplasma or Chlamydia species have been identified in the airways of 45% and 11% of asthmatic subjects, respectively as compared to Mycoplasma in only 9% of normal control subjects(27). A positive relationship between Chlamydia pneumonia-specific secretory-IgA antibody levels and asthma exacerbations(28) provides further evidence for latent bacterial infection impacting asthma severity. In contrast, Sutherland et al(34) identified only 13% of suboptimally controlled asthmatics to have PCR evidence for M. pneumoniae or C. pneumoniae on lower airway endobronchial biopsy. Furthermore, the addition of clarithromycin did not improve asthma control making the clinical relevance of latent atypical bacterial infection unclear. Improved detection methods for atypical respiratory pathogens are needed to more fully characterize the relationship between these pathogens and asthma.
Latent viral infection also appears to be more common among subjects with asthma. Among 50 asymptomatic asthmatic children adenovirus was found in 78.4% of subjects, RV in 32.4%, and coronavirus in 2.7%; co-infection with two or all three viruses was also identified(30). Twenty healthy children were included as controls with only adenovirus detected in one nasopharyngeal swab. Results from other(29) but not all(27;35) studies have supported an increased prevalence of latent adenoviral infection in asthmatics likely due to varying viral detection techniques and small sample sizes. A high incidence of persistent RV infection has also been identified in asthmatics with detectable RV RNA in greater than 40% of asthmatic children 6 weeks after an acute exacerbation(31). RV RNA has also been identified in 73% of mucosal biopsies of asymptomatic patients with asthma as compared to only 22% without asthma, with the presence of human RV significantly associated with lower pulmonary lung function(33). Further studies are needed to investigate the relationship between RV persistence and asthma disease severity. Due to the finding of successive infections in children with different serotypes of human RV(36), studies which incorporate genotyping will also be helpful in determining whether what appears to be a latent RV infection in asthmatics is indeed virus persistence or conversely subsequent infections with different serotypes. This differentiation will be critical to furthering our understanding of RV infection in asthmatics and to guide future therapeutic measures including human RV-specific vaccines.
Due to the strong relationship between RSV bronchiolitis in infancy and the development of asthma(37), latent RSV infection of asthmatics has also been suggested(38). Persistent RSV infection has been documented 100 days post infection in a mouse model(39) and 5 weeks post infection in a guinea pig model(40) but similar findings have yet to be confirmed in human studies, indicating a key area in need of further research. The ability of RSV to infect(41) and persist(42) in vitro within human dendritic cells has been demonstrated and in vivo RSV persistence in human subjects has been suggested by the identification of RNA sequences homologous to the RSV genome in the naïve human bone marrow stromal cells of adult and pediatric donors, but not the complete virus(43).
Perhaps the strongest evidence for a relationship between increased infection susceptibility among persons with asthma and atopic disease is the increased rates of invasive disease (Fig 1). In a study of children with respiratory illness, rhinoviremia was detected in 25% of children presenting with an asthma exacerbation as compared to only 5% of children presenting with other respiratory conditions(44). Among persons 2 to 49 years of age enrolled in Tennessee’s Medicaid Program, the average annual incidence rate of invasive pneumococcal disease among persons with asthma was three-fold higher as compared to persons without asthma (6.1 vs. 2.0 episodes per 10,000)(45). Population-based case-control studies conducted in Minnesota(46) and Finland(47) confirmed these findings, with 17% and 5% of the invasive pneumococcal disease burden attributable to asthma within the respective population studied. An increased risk for serious pneumococcal disease for persons with atopic disease has also been demonstrated(48). Persons with atopic dermatitis are also highly susceptible to widespread disseminated viral infections including eczema mollusculatum, eczema herpeticum, and eczema vaccinatum(49). Interestingly, these conditions are not typically seen in patients with other inflammatory skin conditions, such as psoriasis(50).
A variety of potential mechanisms have been proposed to explain why those with asthma and atopic disease appear to have increased susceptibility to select viral and bacterial pathogens (Fig 2). Impaired innate immune responses have been observed in persons with asthma and atopic disease including deficient epithelial cell function, mucus overproduction, decreased interferon responses, and impaired alveolar macrophage function.
Epithelial cells play important roles in immune responses including maintenance of barrier function, mucociliary clearance, production of peptides which have the ability to kill or neutralize microorganisms, and release of chemokines which influence antigen-specific T and B cells; deficiencies in these abilities would be expected to increase an individual’s susceptibility to infections(51). Examination of bronchial biopsies from asthmatics reveal disrupted tight junctions, an important component of epithelial barrier function, and increased permeability to macromolecules has been demonstrated in epithelial cell cultures(52). A significant association between asthma development and a proteinase inhibitor gene (SPINK5) polymorphism, postulated to impair epithelial barrier function, has also been identified within a population of German children(53). The role of atopy in host susceptibility must also be considered, as allergens have the potential to interfere with proper epithelial barrier function, and the nasal mucosal changes associated with allergic rhinitis are histologically similar to those seen in the lower airways in asthmatics(54). Peptidase allergens disrupt intercellular tight junctions in human bronchial epithelial cell lines, thereby increasing permeability of the airway epithelium(55;56) and the disruption caused by the dust mite allergen, Der p1, enhanced RSV replication within a human bronchial epithelial cell line(57). These studies shed light on how the alteration of the airway epithelium among asthmatic and atopic persons may increase susceptibility to infections.
Mucus production is important in handling respiratory pathogens, and is both increased in quantity and viscosity in asthma. Endobronchial biopsies from asthmatic patients demonstrate airway goblet cell hyperplasia with increased numbers of mucus-secreting goblet cells in the epithelium and an increase in the size of the submucosal glands(58). RV infection in vivo increases release of the major mucin component of airway mucus secretions, MUC5AC, and in asthmatic subjects MUC5AC levels positively correlate with peak virus load(59).
Studies have shown persons with asthma and atopic disease have a deficient interferon response with significantly lower levels reported as early as birth(60), likely contributing to the future risk for respiratory viral infections. An inverse relationship between cord blood IFN-γ responses and the frequency of symptomatic viral respiratory infections has been demonstrated within the first year of life(61). Allergen exposure and RSV both reduce INF-γ production as compared to RSV infection alone in both mouse(62) and rat models(63), and may thus be synergistic in increasing host susceptibility to infection. The importance of the deficient IFN-γ production is further supported by the modulation of postviral sequelae, including significantly less bronchiolar inflammation and fibrosis, when rats with deficient IFN-γ responses receive exogenous IFN-γ supplementation during acute viral illness(64). Deficient IFN-β and IFN-λ responses to infection with RV have been demonstrated in human bronchial epithelial cells in vitro(65;66). Furthermore, the amount of IFN-λ production in bronchoalveolar lavage cells infected in vitro with RV significantly inversely correlated with clinical illness severity of the same subjects infected with RV in vivo(65), although these findings have not been replicated by others. Bochkov et al(67) and Lopez-Souza et al(68) demonstrated no difference in interferon responses in asthmatic bronchial epithelial cells in vitro in response to RV infection as compared to healthy control subjects. More recently in vitro studies in asthmatic nasal epithelial cells have demonstrated lower production of IFN-λ1 post RV infection compared to healthy control epithelium, but higher severity of illness associated with higher levels of IFN-λ1 production by asthmatics in vivo(69). How this relates to human clinical infection risk and illness severity and whether increasing interferon responses would modify morbidity still needs to be elucidated.
Persons with asthma may also have impaired alveolar macrophage function. Airway macrophages from children with moderate and severe poorly controlled asthma were found to have significantly blunted phagocytosis of S. aureus and increased apoptosis(70). Children treated with inhaled corticosteroids were the control group in this study, and while that limits the interpretation, a later study(71) also supported impaired alveolar macrophage function with altered airway and intracellular airway macrophage glutathione homeostasis in children with severe asthma as compared to children with moderate asthma.
Asthma is characterized by enhanced T helper 2 (Th2) activity(72). An increased number of CD4+T cells, predominantly Th2 cells, are seen in the airways of asthmatics, which is in contrast to the T helper 1 (Th1) cell predominance seen in normal airways(73). Delayed postnatal maturation of the immune system, including a delayed transition from Th2 to Th1 bias, is a risk factor for respiratory infections. High IL-5 production by Th2 cells at birth predicts future risk of severe respiratory infections in childhood(74); concomitant IL-10 production by T cells at birth attenuated this risk(74). In a mouse model, after infection of the lungs with Chlamydia muridarum, IL-13, a Th2 cytokine is rapidly produced and promotes susceptibility to infection, possibly related to impairment of macrophage phagocytic function(75). This may explain why allergic asthmatics, with a dominant Th2 response and enhanced IL-13 production, would be more susceptible to Chlamydial lung infection. Down-regulation of toll-like receptors (TLR) may be responsible for increased susceptibility of asthmatics to mycoplasma infection as mycoplasma clearance in an allergic mouse model has been demonstrated to be due to TLR2 downregulation(76). Impaired adaptive immune responses have been described in those with atopic dermatitis as well. Arkwright et al(77) found a significantly lower proportion of children with moderate to severe eczema to have adequate antibody responses to pneumococcal vaccination as compared to control subjects with isolated recurrent upper respiratory tract infections (17% vs. 57%). However, as the children with eczema had no history of severe or recurrent infections with S. pneumoniae, the clinical significance of their reduced response to pneumococcal vaccination is uncertain.
Alterations in immune function may be, at least in part, due to genetic influences as multiple asthma and atopic disease susceptibility genes are functional immune response genes(78;79). The GABRIEL consortium(80), a large genome-wide association study of asthma, identified a significant association for IL18R1, IL33, HLA-DQ, SMAD3, IL2RB, and ORMDL3. Many of these genes have a multitude of functions but all have been found to be involved in the Th2 inflammatory response to epithelial damage sustained during trauma or infection. A variety of genetic mutations have been identified in patients with atopic dermatitis which are associated with skin barrier dysfunction. The most commonly identified is a filaggrin mutation, present in up to 50% of patients with atopic dermatitis(81) and associated with the persistence of atopic dermatitis into adulthood(82) as well as increased asthma severity(83). The downregulation of multiple immune response genes have also been identified in patients with atopic dermatitis(50). It has also been postulated that parental asthma or allergy may influence immune function, with maternal history of atopy or asthma identified as a risk factor for more severe infant HRV-associated illness(84) and lower cytokine responses to innate stimulation by TLR2, TLR3, TLR4 and TLR9 agonists and in vitro stimulation by RSV seen in children with a parental history of allergy or asthma(85).
It has long been debated as to why asthma and atopic diseases have persisted for thousands of years and are prevalent and increasing in most populations. The enhanced Th2 activity seen in asthma and atopic disease may have a teleological basis as it confers a protective advantage against helminthic parasite infection(86-88), far more common pathogens in past centuries(89). Th2 immune signaling in the lungs of asthmatics can have deleterious effects such as increased eosinophil activity, mucus hypersecretion and muscle hyperactivity; however, these same immune mechanisms promote helminth expulsion when expressed in the gut in response to parasitic infection(90). Peisong et al(88) discovered children with upregulated Th2 immune signaling, due to an asthma associated genetic variant of STAT6, had increased resistance to infection with the helminth worm Ascaris lumbricoides. A similar relationship between atopy and helminth infection has been described in Venezuelan children(86) and in an African adult population(87). Together, these studies support a protective advantage against parasitic infections among those with asthma and atopy. It has also been proposed that the reverse is true; that parasite exposure reduces the likelihood of the development of asthma and atopic disease(91).
Other possible mechanisms for increased infection susceptibility include abnormal airway structure and function, due to prior infections and ongoing airway inflammation. Chronic responses to paramyxoviral(92) and RSV infection(93) have been demonstrated in mice consisting of airway hyperreactivity and goblet cell hyperplasia persisting for at least one year after complete viral clearance, and methacholine induced airway hyperresponsiveness up to 154 days post viral inoculation, respectively. Paramyxoviral infection has also been shown to alter epithelial cell function, characterized by decreased airway mucociliary velocity and impaired bacterial clearance(94).
Infection latency or colonization may contribute to enhanced susceptibility to infections by promoting ongoing airway inflammation. In a mouse model, trace levels of parainfluenza virus have been shown to be associated with persistent activation of the NKT cell-macrophage innate immune axis and production of IL-13 leading to chronic mucus cell metaplasia and airway hyperreactivity(95). Long-term persistence of RSV and airway hyperresponsiveness and airway eosinophilia has been demonstrated in the guinea-pig lung(38). These findings are notable, but their clinical relevance remains unclear because despite a likely role for latent parainfluenza and RSV infections in asthmatics, persistence of these viruses in humans has not been clearly established.
Lastly, there are a number of common extrinsic factors that are associated with both increased susceptibility to infection and with asthma exacerbations or asthma control, including cigarette smoke, air pollutants, and nutrition. Maternal smoking is a known risk factor for respiratory infections(96) and the development of asthma(97;98) and atopic disease(99) but has more recently also been shown to attenuate neonatal immune function with impaired TLR-mediated immune responses identified in neonates whose mothers smoked during pregnancy(100). Passive smoke exposure has also been shown in adolescents to be associated with higher rates of pneumococcal colonization of the nasopharynx(23).
Outdoor air pollutants including ozone and diesel exhaust have been implicated to alter immune function and increase infection susceptibility. Exposure to ambient ozone results in the loss of lung epithelial integrity, a key innate immune defense mechanism of the airways(101). Exposure to diesel exhaust increases the susceptibility to influenza virus infection in respiratory epithelial cells both in vitro(102) and in mice in vivo(103), possibly related to the downregulation of antimicrobial host defense molecules, including Clara cell secretory protein and surfactant proteins A and D(104).
Nutritional factors including vitamin D have also been described to impact immune function. Subclinical vitamin D levels are associated with an increased risk for respiratory tract infections in both infants(105;106) and children(107). Proposed mechanisms by which vitamin D levels modulate immune function include induction of antimicrobial peptide expression(108), downregulation of TLR expression by monocytes(109), and inhibition of T and B cell proliferation(110;111). Some of these effects strengthen the immune system whereas others actually suppress immune function, making the exact role of vitamin D in the immune system as well as the ideal circulating levels still unclear. Randomized controlled trials are ongoing to evaluate the role of vitamin D supplementation during pregnancy and asthma development(112;113).
One must also question whether any of the medications used to treat asthma or atopic disease, most notably corticosteroids, modify infection risk. Fortunately, the use of corticosteroids in asthmatics does not appear to be immunosuppressive. The association identified by Talbot et al(45) between invasive pneumococcal disease and asthma remained after adjustment for the long-term use of oral corticosteroids. Furthermore, Wos et al(33) found no relationship between inhaled corticosteroid dose and RV presence in the lower airways of patients with bronchial asthma. Nasal and inhaled corticosteroids may actually be protective in persons with allergic rhinitis or asthma by resulting in restitution of the upper and lower airway epithelium, respectively.
The suggestion of altered host susceptibility to specific viral and bacterial pathogens in persons with asthma and atopic disease has important implications for treatment and prevention. Whereas the current mainstay of therapy is inhaled corticosteroids which inhibit inflammation, more emphasis may need to be placed on therapies which bolster innate and/or adaptive immune responses in response to infection, or in deterring the long-term consequences of infection.
Other therapies which may prove beneficial include increased vaccination efforts, exogenous type I IFN administration, and glutathione supplementation. As a result of the recently demonstrated increased risk of invasive pneumococcal disease among persons with asthma, pneumococcal vaccination is now recommended for all adult asthmatics(114) and has been associated with a decrease in asthma-related hospitalizations and asthma-related length of hospital stay(115). Due to the findings of deficient IFN-β production by asthmatics, Cakebread et al(116) investigated the therapeutic potential of exogenous IFN-β administration on RV infection in asthmatics in primary bronchial epithelial cells from asthmatic donors. Treatment of cells with exogenous IFN-β followed by infection with RV resulted in a normal anti-viral response to RV, including induction of apoptosis and reduced RV replication. Intrabronchial delivery of IFN-α in an asthmatic cohort has not been effective as it induced bronchospasm with a significant decrease in FEV(117)1. Another treatment option under investigation includes the use of glutathione supplementation in patients with severe asthma as these patients have reduced levels of glutathione and subsequent impairment of airway macrophage function. Ex vivo glutathione supplementation has been demonstrated to significantly improve the phagocytic function of airway macrophages collected from severe asthmatics(71).
The relationship between infections and asthma is still not fully understood. There has been historical debate over whether respiratory infections play a causal role, or severe infections are merely a marker of those predisposed to develop asthma. The likelihood is that both are true, hosts who have a genetic propensity to develop asthma, are those who also have an altered immune response to specific pathogens, resulting in more severe infection, and that these early life events are additionally causal in asthma development. The shared pathways demonstrated by recent studies support the possibility of altered susceptibility to specific viral and bacterial pathogens within persons with asthma and atopic disease. More research is needed to characterize the innate and adaptive immune responses in asthma and atopic disease in order to both improve control of these diseases, to develop treatment strategies for those pathogens for which there are currently no therapies, and someday even prevent their development.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.