key: cord-022337-f3a349cb
authors: Busse, William W.; Dick, Elliot C.; Lemanske, Robert F.; Gern, James E.
title: Infections
date: 2007-05-09
journal: Asthma
DOI: 10.1016/b978-012079027-2/50112-x
sha: 
doc_id: 22337
cord_uid: f3a349cb

Wheezing with respiratory infections is extremely common in early childhood. It is estimated that the prevalence of wheezing during the first five years of life varies from 30–60%. In the majority of children who experience wheezing with respiratory infections, these episodes of wheezing become less frequent as the child grows older. However, determining whether the initial episode of wheezing with a viral respiratory illness is an important factor in the eventual development of asthma is an important question. Although a significant body of information suggests an association between respiratory tract illnesses in early life and the later development of airway dysfunction, this relationship is difficult to establish and indicates the complexity of factors that surround the development of bronchial hyperresponsiveness and eventual expression of asthma. A similarly important issue to resolve is the relationship between respiratory infections and the pathogenesis of airway hyperresponsiveness. It is apparent that viral, not bacterial, upper respiratory infections (URIs) trigger asthma attacks. With the use of more sensitive techniques to identify respiratory viruses, the relationship between respiratory infections, particularly viral URIs, and asthma has become even more convincing and important.

these episodes of wheezing become less frequent as the child grows older. However, a question still remains as to whether the initial episode of wheezing with a viral respiratory illness is an important factor in the eventual development of asthma. Although a significant body of information suggests an association between respiratory tract illnesses in early life and the later development of airway dysfunction, this relationship is difficult to establish and indicates the complexity of factors that surround the development of bronchial hyperresponsiveness and eventual expression of asthma. Furthermore, the source of subjects for study, i.e. follow-up of hospitalized patients vs. outpatients, contributes to the difficulty of understanding this problem.

Eisen and BacaP found that children hospitalized for bronchiolitis prior to age 2 years had an increased risk for asthma. Rooney and Williams'^ also evaluated, retrospectively, the records of infants hospitalized for bronchiolitis at 18 months or younger; allergic manifestations and a family history of asthma were more frequent in children who eventually experienced one or more episodes of wheezing. Finally, McConnochie and Roghmann^ identified 77 patients who had bronchiolitis at 25 months or younger and compared their outcome to children without a history of bronchiolitis. When these children were evaluated approximately 7 years later, only upper respiratory allergy, bronchiolitis and passive smoking exposure were found to be independent predictors of wheezing following bronchiolitis. Consequently, it is apparent that the final conclusions on the relationship between respiratory infections in infancy and later asthma must consider a host of influences, including parental smoking, underlying airway responsiveness and gender.

A similarly important issue to resolve is the relationship between respiratory infections and the pathogenesis of airway hyperresponsiveness. T o evaluate the effect of bronchiohtis on airway responsiveness, Sims et al.^ identified 8-year-old children who had respiratory syncytial virus (RSV) respiratory infections and quantitated bronchial 'lability' by exercise tests. Compared with appropriate controls, the fall in the peak flow with exercise was greater in children who had bronchiolitis; however, airway reactivity to exercise was not different between children with or without subsequent episodes of wheezing. Since other variables confounded their study, Sims et al.^ could not prove that respiratory infections led to the later development of asthma.

Other efforts have been made to ascertain if viral lower respiratory tract infections (LRIs) in early life cause persistent pulmonary function abnormalities. Pullan and Hey^ evaluated 130 children admitted to hospital during the first 5 years of life with RSV LRIs; 42% of the hospitalized children had future episodes of wheezing, while only 19% of control subjects experienced similar airway symptoms. However, few patients (6.2% vs. 4.5% of controls) had troublesome respiratory symptoms by 10 years of age. Furthermore, although a three-fold increase in bronchial responsiveness was found in the children with bronchiolitis, atopy was not increased. Analogous conclusions were reached by Weiss et al^ when they assessed the outcome of an antecedent acute respiratory illness on airway responsiveness and atopy in young adults. Airway responsiveness, evaluated by eucapnic hyperventilation to subfreezing air, was increased in children with a previous history of either croup or bronchioHtis, or greater than two acute lower respiratory illnesses.

The possibility has also been raised that a predisposition to wheezing in infancy depends more on intrinsic airway structure than atopy.^ This position is supported by the high degree of airway responsiveness found in infancy in physiological evaluations^^~^^ and the incidence of wheezing with respiratory infections.^^ However, it is difficult to precisely assess airway responsiveness in young children due to limitation of lung size and other age-related factors.

To help clarify the relationship between premorbid lung function and wheezing with respiratory illnesses, Martinez et al}^ conducted a prospective study of respiratory illness in infancy and childhood. Lung function values were determined prior to any LRIs. Included in these measurements were tidal expiratory patterns, specifically the time to peak tidal expiratory flow (Tme) divided by total expiratory time ( T E ) , or the Tme/TE ratio; Morris and Lane^^ had shown that decreasing Tme/TE ratios correlated with lower lung function in patients with progressive chronic obstructive lung disease.

Of the infants studied by Martinez et al,^^ 36 developed an LRI and 24 wheezed with at least one of these infections. There was no diflerence in preinfection lung function between those infants who did not have an LRI and those with an infection but no wheezing (Table 30 .1). However, infants who wheezed with the respiratory infection had diminished Tme/TE values and reduced expiratory system conductance when measured prior to wheezing with the infection. These data suggest that alterations in lung function are compatible with reduced airway conductance or a slow respiratory system time constant that precedes and predicts wheezing with respiratory infections in infants. Furthermore, it appears that a given child's response to infection is determined not only by the infection but also by pre-existing lung function.

Taussig et al}^ also noted that lower levels of lung function predispose to wheezing with LRI, as opposed to the infection per se. The precise nature of this predisposition remains to be defined but may lie in airway geometry, airway-parenchymal interaction, or mucosal and smooth muscle response. Furthermore, this pulmonary-structural predisposition may be enhanced by an exaggerated IgE response to viral infection,^^'^^ resulting in more inflammation and severe wheezing with hospitalization. Since the majority of infants who develop wheezing with LRIs do not wheeze throughout life,^^ it is likely that pulmonary function abnormalities that favour wheezing with viral infections are modified with the growth and development of the lung. Long-term outcome then seems to be more closely linked to the persistence of ongoing airway damage or bronchospasm associated with the development of atopy and true clinical asthma.

Further, and possibly definitive, insight into the relationship between wheezing with early respiratory infections and the later development of asthma has come from a unique They identified a number of factors that affect wheezing before the age of 3 years and their relationship to wheezing at 6 years of age. The study population has previously been reported^"^ and consists of newborns enrolled between 1980 and 1984 with follow-up information at 3 and 6 years of age. Key assessments in infancy included cord-serum I g E levels, pulmonary function testing before any lower respiratory tract illness had occurred, measurement of serum I g E at 9 months of age, and a questionnaire completed by the children's parents when the child was 1 year old. The children were classified into four groups: no wheezing, transient wheezing, late-onset wheezing or persistent wheezing (Table 30 .2, opposite). At 6 years of age, serum IgE, pulmonary function testing and allergy skin testing were repeated and these factors were assessed in relationship to their history of wheezing. At 6 years of age, 20% of the children had at least one lower respiratory illness with wheezing during the first 3 years of life, but with no wheezing at age 6 years. These children had diminished airway function before the age of 1 year and, at 6 years of age, were more likely to have mothers who smoked but not mothers with asthma and did not have evidence of atopy, i.e. elevated serum I g E or skin test reactivity. In addition, 15% had no wheezing before the age of 3 years but had wheezing at age 6 years and 13.7% had wheezing both before 3 years of age and at 6 years of age. Those with late-onset wheezing and persistent wheezing were more likely to have mothers with a history of asthma, elevated serum I g E levels and diminished lung function at 6 years of age (Tables 30.2 

-M G G <u g^ <u 0 G ^ qj ON rG G i n c ( N (J <-i H i j 0 <u B o O <U < ! oj G toe G ' N (U <u (U w G O <u - § 1 G <u 2 la o <u ^ •^ CO G 0 CO a 0 u VH o i n o d V GH G c3 TH (U N <U (U VH > G 0 G <u 4-1 G C/3 G O CO ' v H a. a o u a c« "G u G :3 CO ' -' a ^ 0 u VH o c/^ O G N (U <U •»-> G CO <U G 0 CO "v1 a.

A number of conclusions can be drawn from this prospective longitudinal study. Most infants who wheeze early in life have a transient condition and this predilection to wheezing with respiratory infection is associated with diminished lung function; these children are not at enhanced risk for asthma or allergic disease later in life. The groups of particular interest regarding the role of respiratory infections and development of asthma are those infants with either persistent or late-onset wheezing. Children with asthma at age 6 years appear to have a predisposition to asthma, as indicated by a maternal history of asthma and elevated serum I g E levels in the first years of life. Moreover, these children have reduced lung function by 6 years of age. The role of respiratory infections in the actual development of asthma in those with a predisposition to asthma is less clear; that is, do respiratory infections have any relationship to the subsequent development of asthma? This issue remains to be fully elucidated. Insight to factors that may determine a host's response to viral infections and its subsequent effects on lung function has been gained by animal studies. Sorkness et al?^ infected neonatal rats with Sendai virus. At 7 and 13-16 weeks afer infection, the rats that had been infected had lower P02 values and increased lung resistance when compared with controls. Furthermore, the infected animals were more sensitive to aerosolized methacholine. Lung histology in the infected animals revealed increased numbers of bronchiolar mast cells and eosinophils. In subsequent studies, these investigators have found some rat strains, i.e. Brown Norway, are more susceptible to the effects of the virus and more likely to have persistently altered lung function. Such studies in an animal model are compatible with the observations by Martinez et al?^ and indicate that many factors determine the eventual influence of respiratory infections on airway function, i.e. wheezing and bronchial hyperresponsiveness, including a predisposition to produce IgE antibodies. However, what needs to be clarified are the mechanisms by which respiratory viruses interact with IgE-dependent factors and promote an asthma-like picture.

For decades clinicians associated asthma exacerbations with respiratory infections. With prospective studies, it became apparent that viral, not bacterial, upper respiratory infections (URIs) triggered these asthma attacks. With the use of more sensitive techniques to identify respiratory viruses, the relationship between respiratory infections, particularly viral URIs, and asthma has become even more convincing and important.

In early studies, Mcintosh and coworkers^^ prospectively studied 32 young children, aged 1-5 years, with severe asthma. The aetiology of each URI was carefully evaluated and confirmed by culture and serological tests for viral antibody titres. Over 2 years, there were 102 confirmed viral respiratory infections and 139 episodes of wheezing; 58 episodes (42%) of wheezing occurred in relationship to viral respiratory infections, of which RSV was the most prevalent and likely to provoke asthma. Respiratory bacteria, Haemophilus influen:(ae. Streptococcus pneumoniae, j?-haemolytic streptococcus and Staphylococcus aureus, were also cultured but their presence did not correlate with an asthma attack.

A prospective outpatient study from the University of Wisconsin evaluated 16 children, aged 3-11 years, with histories of four or more asthma attacks associated with respiratory illnesses during the previous year.^^ Detailed clinical records profiled each child's asthma severity, which was further substantiated with biweekly examinations. During an asthma exacerbation or apparent URI, additional bacteria and virus cultures were collected and asthma symptoms carefully quantitated. The 16 children experienced 61 episodes of asthma; 42 occurred in conjunction with a symptomatic respiratory infection and 24 were confirmed to be of viral aetiology by culture and/or serum haemagglutination titres. In this study, rhinovirus was the most frequently identified virus in association with wheezing. Some patients had episodes of asymptomatic viral infection, but asthma was not worsened. Only one episode of wheezing coincided with a bacterial infection.

To extend these observations and to further determine if there was a relationship between different respiratory infections and wheezing, Mertsola et al?^ identified 54 patients, aged 1-6 years, who had recurrent attacks of wheezing. Over a 3-month observation period, these patients experienced 115 episodes of upper or lower respiratory symptoms. Of these episodes, laboratory evidence of a viral or Mycoplasma infection was found in 45%. More relevant to our discussion was the observation that wheezing occurred in 58% of the laboratory-confirmed viral respiratory infections. Moreover, of the children with wheezing with respiratory infections, one-third were admitted to hospital because of severe dyspnoea. Finally, of the agents isolated, coronavirus and rhinovirus were the most frequent isolates.

There is also evidence that the viruses which cause wheezing with respiratory tract infections may be age dependent; for example, infants wheeze with RSV while older children have exacerbations of asthma with rhinovirus.^^ To extend these observations. Duff and colleagues^^ examined the relationship of viral infections, passive smoke exposure and IgE antibody to inhaled allergens in infants and children who were treated for acute episodes of wheezing in an emergency department. The investigators identified 99 subjects, aged 2 months to 16 years of age, and 57 control patients, aged 6 months to 16 years of age, who were seen in a paediatric emergency room. In acutely wheezing children less than 2 years of age, viruses were detected by culture of nasal washings in 70% of the patients; the most commonly identified virus was RSV. Positive virus cultures for RSV were noted in only 20% of the appropriate age-matched controls. In contrast, nasal washes yielded positive cultures in only 31% of children over 2 years of age who presented with acute wheezing. However, the most frequently identified respiratory virus in this age group was rhinovirus.

When these investigators evaluated risk factors associated with wheezing during the viral infections, interesting and insightful correlates were noted (Table 30 .4). For children less than 2 years of age, coexisting allergic diseases were not a risk factor. Rather, passive exposure to cigarette smoke, as indicated by urinary cotinine values ^1 0 jUg/ml, was a risk factor for wheezing. The profile in children over 2 years of age with wheezing and viral URIs was quite different. In these children, the presence of allergy, i.e. positive radioallergosorbant test (RAST) values, and not passive cigarette smoke exposure, was the dominant risk factor for wheezing with viral respiratory infections. Observations by Duff et al}^ confirm previous information that RSV is the major viral infection causing wheezing in infants, whereas rhinovirus is the causative respiratory infection associated with wheezing in older children. Moreover, wheezing in infants is not increased by the presence of allergic disease but rather made more probable if cigarette smoke exposure is present. In the older child, allergies are a major risk factor for wheezing with rhinovirus; these data also imply that a unique relationship may exist between rhinovirus and allergy, and this interaction may influence the development of wheezing with this particular respiratory virus. The detection of respiratory viruses, particularly rhinovirus, by culture is difficult because of the fastidiousness of viruses in culture. Consequently, determination of the association between rhinovirus infections and provocation of asthma using virus culture may give an underestimation of the true frequency between colds and asthma. Johnston ef al}^ developed a polymerase chain reaction (PCR) assay to detect picornavirus infections. This approach has an increased sensitivity and thus greater likelihood to detect causative viruses in relationship to clinical symptoms of a cold and eventual influence on asthma.

Using PCR technology, along with culture, Johnston and colleagues^^ studied the association between upper and lower respiratory viral infections and acute exacerbations of asthma in school-age children; 108 children, 9-11 years of age, were followed over a 13-month period. Children or their parents recorded the peak expiratory flow rate twice daily and daily upper and lower respiratory tract symptoms were scored as to severity. With the onset of respiratory symptoms, or a significant fall in peak expiratory flow values, the child was visited by an investigator and nasal aspirates were obtained for virus detection.

There were a number of important observations by Johnston and colleagues in this large study. First, there was a seasonal pattern to the appearance of respiratory symptoms and changes in peak flow values of the asthma patients; the months of N o v e m b e r -December and April appeared to be the most frequent times for colds and asthma exacerbations. Second, respiratory viruses were detected in approximately 80% of episodes with respiratory tract symptoms or falls in the peak flow (Table 30 .5). Third, rhinovirus was the most frequently detected microorganism and constituted over 60% of the viruses detected (Table 30, 6) . Fourth, the severity of asthma symptoms in many of these subjects was severe. From these observations, and those of others, it is apparent that rhinovirus, the cause of the common cold, is the respiratory infection most likely to exacerbate wheezing in individuals with existing asthma. It is possible that this relationship simply reflects the fact that rhinovirus is the most common cause of infectious respiratory symptoms, i.e. the common cold. However, this association may also mean In contrast to observations in children, the relationship between respiratory infections and episodes of wheezing is not as striking in adults. When both children and adults were evaluated, respiratory viruses were more frequently identified during episodes of asthma in children under 10 years of age compared with adults. Eight adult subjects evaluated by Minor et alP had only three documented virus infections with increased wheezing. Although an explanation for such findings is not clear, it is likely that asthma flares were more diflficult to sharply delineate and the pattern of wheezing often appeared chronic rather than episodic in adults. Similarly, when Hudgel and coworkers^^ evaluated 19 adult asthma patients over a 15-month period, 76 episodes of asthma were recorded but only eight could be documented with a viral URL Although clinically apparent viral respiratory infections precipitate asthma in adults, the relative frequency appears less than in children. In another study, Huhti et al?^ evaluated adults admitted to hospital with asthma exacerbations. The viral detection rate was 19.0%, with RSV, parainfluenza and influenza the more frequent isolates.

Using virus culture and PCR techniques, Nicholson and coworkers^^ evaluated the relationship of symptomatic colds and asthma exacerbations in 138 adult asthmatic subjects. These investigators found symptomatic colds were reported in 80% of episodes with symptoms of wheeze, chest tightness or breathlessness. In 24%) of laboratoryproved episodes of non-bacterial respiratory infections, there was a reduction in peak flow ^ 50 litre/min. Infections with rhinoviruses, coronaviruses, influenza, RSV, parainfluenza and Chlamydia were all associated with objective measures of airflow limitation. The findings of Nicholson et al. are further evidence of the importance of respiratory viruses, particularly rhinovirus, to asthma exacerbations, and that these episodes can be severe and occur in adults. Therefore, evidence exists to underscore the importance of viral respiratory infections to asthma exacerbations in patients of all ages.

There is little evidence that bacterial respiratory infections, other than sinusitis, provoke asthma. Berman et al?^ performed transtracheal aspirates on 27 adult patients with asthma during an infectious exacerbation. Bacteria cultured from transtracheal aspirates were sparse and, more importantly, did not correlate with clinical illness. Moreover, transtracheal aspirates from normal subjects, without respiratory infections, yielded a similar degree of bacterial colonization. A common complication of colds is the development of sinusitis, which has been proposed as a factor in the worsening of asthma. Although the association between paranasal sinusitis and bronchial asthma has long been observed, it is conceivable that sinusitis and asthma may coexist as complications of the same respiratory infection. However, some feel that an aetiological relationship exists between sinusitis and asthma.^"^'-^^ Although these observations and associations are intriguing and clinically important, additional studies are needed to clarify how sinusitis may influence asthma.

A number of mechanisms have been proposed to explain how viral respiratory infections provoke asthma (Table 30 .7). When evaluating these possibilities, it is important to realize that the effect respiratory viruses have on airway function will depend upon many factors, including the subject's gender, environmental factors such as cigarette smoke exposure, age, baseline lung function and atopic status, along with the particular respiratory virus. However, there is now convincing evidence that rhinoviruses are the predominant respiratory infection provoking wheezing in patients with asthma and the existence of allergic disease appears to be an important risk factor. Therefore, it is proposed that respiratory viruses, particularly rhinoviruses, may interact with those factors that contribute to or determine existing allergic airway disease and enhance existing bronchial inflammation. The end-result of the eflects of the virus on underlying 

Altered epithelial function Altered autonomic nervous system regulation of airway smooth muscle function Enhanced allergic inflammation Enhanced eosinophilic recruitment Eosinophilic infiltration of the airway Generation of pro-inflammatory cytokines Enhanced inflammatory mediator release Altered j5-adrenergic function Sinopulmonary reflex Production of virus-specific IgE antibodies allergic airway disease will be an enhanced likelihood for wheezing. Although other factors likely participate in this process, the following discussion examines how respiratory viruses either influence IgE-dependent events or promote the development of allergic inflammation.

Welliver et al}'^ tested 79 children, all less than 12 months of age and with documented RSV infection, for the presence of IgE-specific antibody to the infecting virus. Their clinical patterns of illness were divided into four groups: (a) upper respiratory tract illness; (b) pneumonia without wheezing; (c) pneumonia and wheezing; and (d) wheezing (bronchiolitis). Nasal secretions from each patient were measured for IgE-specific antibody to RSV, and RSV IgE antibody titres were compared with the patient's clinical illness. I g E titres to RSV were highest in patients with evidence of airway obstruction, e.g. pneumonia/wheezing or bronchiolitis. Parainfluenza virus-specific IgE responses were also examined in individuals with upper respiratory illness alone or bronchiolitis by Welliver et al}^ Parainfluenza-specific IgE antibody were detected in 8 of 12 with bronchiolitis but only 1 of 10 with upper respiratory disease alone. These studies suggest that some respiratory viruses stimulate IgE-specific antibody responses and the degree to which this occurs is associated with the clinical manifestations of upper vs. lower airway disease.

To further evaluate the significance of virus-specific I g E antibody, Welliver et al?^ prospectively monitored 38 infants for 48 months after an initial episode of bronchiolitis. Only 20% of infants with undetectable titres of RSV IgE had subsequent episodes of documented wheezing. In contrast, 70% of those children with high RSV I g E antibody titres continued to experience wheezing when evaluated 4 years later. A portion of these same subjects was evaluated at 7-8 years of age^'^ and underwent skin testing to common inhalant allergens, pulmonary function testing and methacholine provocation to determine bronchial responsiveness. For the entire study groups, RSV I g E responses were evaluated in relationship to family history of asthma, passive smoke exposure, number of recurrent wheezing episodes, number of positive skin tests and the degree of airway responsiveness to methacholine. The investigators found that an association exists between RSV IgE responses and the development of any wheezing following bronchiolitis; the nature, however, of the association between RSV I g E antibodies and any recurrent wheezing is unclear. Therefore, the consequence of an initial I g E response to the infecting virus with regard to long-term influence on airway function and the development of asthma is unresolved.

Although rhinoviruses have emerged as the most common respiratory infection causing asthma exacerbations,^^ there are a number of paradoxes to this relationship. First, rhinoviruses are infections that appear localized to the upper airway; identification of rhinovirus by culture in the lower airway is unusuaP^ and pneumonic infiltrates rare. Second, even though rhinoviruses infect airway epithelium, the inflammatory response is often unimpressive. For example, Fraenkel and coworkers^^ performed nasal biopsies before and after an experimental rhinovirus infection. In both normal and atopic groups, there were no significant changes in inflammatory cells during the cold or the convalescent period compared with baseline values. Thus, immunohistochemical analysis of rhinovirus-infected upper airways does not reveal inflammation. Third, rhinovirus infections tend to be self-limited and later sequelae are rare. Therefore, the mechanisms of lower airway dysfunction are not readily explained by an overzealous inflammatory airway response to this particular virus.

The frequent association of asthma with respiratory infections, and in particular rhinovirus illnesses, may relate to the frequency with which rhinoviruses are the cause of colds. This frequency, however, does not explain the apparently serious lower airway events that are associated with rhinovirus infections in subjects with asthma. A major goal of current research is to establish how this particular class of respiratory viruses affects lower airways so profoundly and how these effects can last for weeks beyond the acute respiratory illness.

In an initial series of experiments to evaluate the effects of respiratory infection on airway responsiveness, we infected subjects with rhinovirus experimentally and determined its effect on a variety of airway functions."^^ The patients selected for study were evaluated on three separate occasions: at baseline, during acute infection and during recovery. Of particular interest was the effect of the infection on the immediate and latephase allergic response (LAR) to inhaled antigen. All 10 patients had a rhinovirus respiratory infection at time of study. During the acute rhinovirus respiratory infection, airway responsiveness to histamine significantly increased over baseline values. Likewise, acute airway reactivity to inhaled antigen was increased. The change in airway responsiveness to histamine and antigen was similar, suggesting that the effects of the respiratory infection on responsiveness to antigen related to alterations in non-specific bronchial reactivity.

Prior to rhinovirus inoculation, only 1 of 10 patients had an LAR to inhaled antigen. However, during the acute respiratory infection, 8 of 10 patients experienced late-phase airway obstruction to inhaled antigen challenge. Furthermore, when evaluated during the recovery period (4 weeks after rhinovirus inoculation), five of the seven patients available for testing still had LARs to inhaled antigen. These observations indicate that rhinovirus respiratory infection not only increased airway responsiveness but also changed the pattern of the airway response to inhaled antigen and promoted the development of allergic inflammation, as indicated by the development of the LAR.

The recovery pattern of airway hyperresponsiveness following the viral respiratory infection was also evaluated. Although increased airway responsiveness was still detected 4 weeks after rhinovirus inoculation, there was a trend towards recovery in both the reaction to inhaled histamine and the response to antigen. However, as already discussed, the increased frequency of LARs to antigen was still noted 4 weeks after the viral infection, suggesting that a viral respiratory infection has a greater, and possibly more lasting, effect on factors that participate in the development of LARs.

Virus-associated airway hyperresponsiveness is a multifactorial process involving a complex interplay of IgE-dependent reactions, epithelial activation or damage, autonomic nervous system dysfunction and, of particular interest and relevance to our discussion, enhanced allergic inflammation."^^ In IgE-mediated reactions, the tissue response, be it the skin, nose or airway, is influenced by I g E sensitization of mast cells and basophils, release of bronchospastic and inflammatory mediators from sensitized cells, and the response of the target organ, which in asthma is bronchial smooth muscle. T o evaluate the effects of respiratory viruses on a facet of the immediate hypersensitivity response, i.e. mediator release, basophil histamine secretion was determined. In an in vitro model system, peripheral blood mononuclear cells were incubated with influenza A virus and the effect of this exposure upon IgE-dependent basophil histamine release and leukotriene (LT)C4 was measured. Following incubation with virus, both histamine and LTC4 secretion was enhanced. T o define the role of Tcells in this process, T-cells were removed by magnetic head separation with anti-CD3 antibody. In the absence of T-cells, virus incubation did not enhance histamine release."^^ Furthermore, analysis of virus-treated peripheral blood mononuclear cells revealed that the virus caused the generation of interferon (IFN)-y. These findings suggest that T-cells, and their cytokine products such as IFN-y, may play an integral role in the processes by which respiratory viruses enhance basophil histamine release, and are likely to affect other cell functions.

The use of bronchoscopy with lavage and biopsy has provided direct assessment and analysis of airway mediators and the histology of inflammatory events in asthma. We have used fibreoptic bronchoscopy to perform segmental bronchoprovocation with antigen and bronchoalveolar lavage (BAL) to assess immediate and late responses of the airway to antigen and to study the effects of a rhinovirus illness on these allergic events."^"^ T o conduct these experiments, a bronchoscope is introduced into the airways, bronchial segments are identified and an allergen is administered directly on to the airway mucosa. The immediate response to antigen is characterized by a brisk rise in mast cell mediators. When lavage of the previously allergen-challenged segments is performed 48 h later, there is a highly cellular, eosinophil-rich response detected. Based on our previous observations that a rhinovirus 16 infection increases bronchial responsiveness and LARs to inhaled antigen, we used segmental antigen challenge and lavage to test our hypothesis that the rhinovirus upper respiratory illness may enhance the lower airways allergic inflammatory response to antigen. Histamine is released into the airway immediately following antigen challenge and represents mast cell activation. We found histamine release into the airway immediately following local antigen challenge to be significantly potentiated by rhinovirus infection (Table 30 .8). These observations indicate that an acute rhinovirus illness can enhance airway mast cell mediator release. These findings are similar to earlier reports and indicate an effect of the infection on mast cell activation."^"^ Interestingly, and relevant to earlier observations that rhinovirus effects on airway function are long-lasting, enhanced histamine release to antigen was still present 6 weeks after the acute infection; at this time, the individuals were symptom-free. Our findings suggest that viral effects on allergic responses can persist for weeks beyond the acute illness and that such persistence can explain the long-lasting effects on airway function in asthma from a cold.

When lavage is performed 48 h after antigen challenge, there is a characteristic and often dramatic influx of eosinophils into the lavage fluid. We found significant enhancement of eosinophil recruitment to the airway 48 h after antigen when the subjects had an acute rhinovirus infection (Fig. 30.1) . Like the observations with enhanced and persistent histamine release, not only was a greater eosinophil recruitment to antigen challenge noted during the acute infection, but this respiratory virus effect persisted into the recovery period 6 weeks later. These findings suggest a possible explanation for the increase in LARs to antigen noted during acute rhinovirus infection; if allergen exposure during rhinovirus infection is associated with enhanced eosinophil recruitment to the airway, allergic inflammation and an LAR are more likely to occur.

Fraenkel and colleagues'^^ provide additional insight into the mechanisms by which rhinovirus infections promote airway inflammation and the likelihood for increased asthma during colds. Both normal and asthmatic subjects were recruited for their study. The investigators obtained bronchial mucosal biopsies before and during an acute experimental rhinovirus infection and evaluated the changes in inflammatory cell infiltrate due to the virus and airway responsiveness. An increase in the airway response to histamine during the cold was associated with increases in submucosal lymphocytes. Further, there was an increase in epithelial eosinophils during the experimental cold ( Fig.  30.2) . In asthmatic compared with normal subjects, the increase in mucosal eosinophils persisted into the convalescent period. Although the changes in various features of immunohistology were not dramatic, they did follow a pattern that is consistent with our findings of enhanced eosinophil recruitment to the airways and were correlated with alterations in lung physiosiveness. Collectively, there is evidence from airway physiology that rhinovirus infections enhance bronchial responsiveness and histological features of allergic inflammation, i.e. there is a greater likelihood for LARs to occur. Furthermore, from studies that directly sample the lower airway, there is evidence of increased eosinophil recruitment and mucosal eosinophilia at the time of rhinovirus infection, and these abnormalities persist for weeks beyond the acute illness. All these findings indicate that rhinovirus infections can perpetuate features of allergic inflammation and may thus explain an increase in the likelihood of asthma.

The questions now remaining to be answered concern the mechanisms by which rhinovirus infections enhance allergic airway inflammation. Using the nasal mucosal response to rhinovirus as a surrogate of the airway response to rhinovirus, Fraenkel et al?^ biopsied nasal tissue during an acute rhinovirus infection. Mast cells, eosinophils, lymphocytes and neutrophils were identified by appropriate monoclonal antibodies. There were no significant changes in the numbers of inflammatory cells present during the cold or the convalescent period compared with baseline values. Although these observations contrast with biopsy findings in the lower airway,'^^ changes in airway 0.048 0028

100-, histology were not dramatic. These findings suggest that other mechanisms, i.e. enhanced mediator release or, more likely, cytokine generation, are responsible for changes in airway function during a cold.

In our study with segmental antigen challenge,"^-^ the lavage fluid was analysed for tumour necrosis factor (TNF)-a. Lavage samples obtained 48 h after antigen challenge had a significant rise in TNF-a; these changes were noted in both normal controls and allergic subjects. O u r findings raise the possibility that rhinovirus infection causes secretion of TNF-a and if enhanced secretion of cytokines were to occur in the allergic subject, features of allergic inflammation are more likely to occur. These findings also raise the possibility that rhinovirus can directly stimulate cytokine production. Linden et a/.^^ measured IFN-y, interleukin (IL)-1)S, granulocyte-macrophage colony-stimulating factor (GM-CSF), IL-4 and IL-6 in nasal secretions of subjects with allergic rhinitis. During an experimental coronavirus cold, there was an increase in IFN-y but not the other cytokines. The increase in IFN-y is expected and represents an antiviral response to the virus. However, IFN-y can have pro-inflammatory effects, i.e. promotion of adhesion protein expression and enhanced survival of eosinophils."^^ Moreover, these observations indicate that rhinovirus infections can stimulate cytokine production in vivo.

T o extend these observations, live rhinovirus was incubated with human mononuclear cells and lung macrophages. Gern et al.^'^ found that these cells took up the virus, but it did not replicate inside the cells. None the less, rhinovirus was able to activate monocytes and airway macrophages and cause secretion of TNF-a. These observations indicate that rhinovirus can cause synthesis and release of pro-inflammatory cytokines, which may occur independent of cell infection.

To identify the mechanisms by which rhinovirus promotes allergic inflammation, it is essential that the immune response to this virus be established so that the factors, and cells, involved in the promotion of allergic inflammation are identified. In determining the effect of rhinovirus on the immune response, it has been established that the receptor for the majority (90%) of rhinovirus strains is intercellular adhesion molecule (ICAM)-!."^^ The other 10% of rhinovirus strains use the low-density lipoprotein receptor.^^ Interaction of rhinovirus with these receptors is the first step in the host's response to this virus. With this information, we have found that a major rhinovirus, rhinovirus 16, binds to mononuclear cells via the ICAM-1 receptor, with binding occurring predominantly to monocytes but not lymphocytes. When rhinovirus 16 was incubated with a mixture of monocytes and lymphocytes, the following events occurred: there was a large increase in the number of CD3 "^ cells that express the early activation marker CD69; when the CD3 "^ /CD69 "^ cells were isolated and analysed, they were found to have increased expression of mRNA for IFN-y. These in vitro observations suggest that rhinovirus activates monocytes to generate cytokines that stimulate lymphocytes. Once activated, these lymphocytes (CDS "^/CD69 "^) generate a variety of cytokines, including IFN-y. Although IFN-y is normally important in antiviral activity, it also possesses a number of pro-inflammatory functions, including increased expression of endothelial and epithelial ICAM-1 (more receptors for rhinovirus) and the promotion of eosinophil survival. Therefore, in addition to activation of a small number of rhinovirus-specific lymphocytes, rhinovirus can cause a 'non-specific' activation of lymphocytes through the generation of cytokines from monocytes; these cytokines can, in turn, promote either viral elimination or inflammation. Perhaps the sequelae of this particular response are more likely if allergic inflammation already exists.

To extend this possibility, Einarsson and coworkers^^ evaluated the effect of respiratory viruses on stromal cell production of IL-11. IL-11 is an extremely cationic cytokine and has the capacity to induce airway hyperresponsiveness in the murine lung. These investigators found that rhinoviruses, along with RSV and parainfluenza, were potent stimulators of IL-11. In contrast, cytomegalovirus and adenovirus were weaker activators and are viruses not associated with asthma exacerbations. Furthermore, increased levels of IL-11 were found in nasal secretions of children with URIs and were more likely to appear in the nasal aspirates of those children with reactive airway disease. These observations are another example of the mechanism by which respiratory viruses can enhance airway inflammation through the generation of specific cytokines and thus increase the likelihood of asthma during a cold. Using a mouse model, Coyle et al?^ characterized the immune response to virus peptides and the resulting cytokine patterns. These investigators found that bystander CD4"^ Th2 immune responses to ovalbumin switched peptide-specific CDS"^ T-cells in the lung to IL-5 producers. When these IL-5-producing CDS"^ T-cells were challenged, via the airways, with virus peptide, a significant eosinophil infiltration was induced. Furthermore, in vitro studies indicated that IL-4 could switch virus-specific CD8~^ T-cells to IL-5 production. These results demonstrate a network by which allergic eosinophilic inflammation can occur during a respiratory viral infection and how this mechanism would impair CD8~^ T-cell responses (less IFN-y production) and thus delay virus clearance. Such observations indicate not only a network to promote allergic inflammation but also a mechanism by which virus particles may persist for extended periods.

Our studies in humans, and those of Coyle et al}^ in the mouse, suggest a possible link between eosinophils and airway changes during viral respiratory infections. When Garofalo et alP sampled nasolaryngeal samples from children with RSV respiratory illnesses, the eosinophil protein, eosinophil cationic protein, was significantly increased in samples from children with bronchiolitis when compared with subjects with upper airway disease or RSV pneumonia but no wheezing. These data suggest that eosinophil activation by RSV may cause airway injury and the presence of wheezing. To extend these findings with in vitro experiments, Kimpen et al}"^ incubated isolated human eosinophils with RSV. RSV directly activated eosinophils to release superoxide and LTC4. Furthermore, incubation of eosinophils with RSV primed the cell for further activation and generation of inflammatory mediators. Both observations indicate that an airway-active virus, in this case RSV, can interact with eosinophils and cause them to release proinflammatory mediators. As indicated by the authors, these effects of the virus could play a role in the pathogenesis of RSV bronchiolitis and may also be important for the development of more long-term bronchial hyperresponsiveness associated with viral infections. It is also possible that other respiratory viruses, such as rhino virus, may have similar effects during a cold.

The mechanisms involved in the development of airway hyperresponsiveness, airway obstruction and recurrent wheezing with viral respiratory infections have yet to be fully appreciated. None the less, current evidence indicates that viruses, or products of virusinfected cells, influence the inflammatory property and potential of many cells. Precisely how these effects of the virus translate into increased airway injury, responsiveness and obstruction will require further work. As the mechanisms of these interactions are established, so will improved understanding of asthma pathogenesis and treatment.

The significance of early wheezing

Lung function, pre-and post-natal smoke exposure, and wheezing in the first year of life

The relationship of acute bronchioHtis to bronchial asthma: a 4-to-14 year follow-up

The relationship between proven viral bronchiolitis and subsequent wheezing

Bronchiolitis as a possible cause of wheezing in childhood

Study of 8-year-old children with a history of respiratory syncytial virus bronchiolitis in infancy

Wheezing, asthma, and pulmonary dysfunction 10 years after infection with respiratory syncytial virus in infancy

The relationship of respiratory infections in early childhood to the occurrence of increased levels of bronchial responsiveness and atopy

Viral respiratory infection in infancy: provocation or propagation? Semin Respir M^^

Airway reactivity in infants: a positive response to methacholine and metaproterenol

Airway response to cold, dry air in normal infants

Response of normal infants to inhaled histamine

\ The Tucson children's respiratory study. 11. Lower respiratory tract illness in the first year of life

Diminished lung function as a predisposing factor for wheezing respiratory illnesses in infants

Tidal expiratory flow patterns in air-flow obstruction

Lung function in infants and young children: functional residual capacity, tidal volume, and respiratory rate

The development of respiratory syncytial virus specific IgE and the release of histamine in nasopharyngeal secretions after infection

Role of parainfluenza virus-specific IgE in pathogenesis of croup and wheezing subsequent to infection

Lower respiratory illness in early childhood and lung function and bronchial reactivity in adolescent males

Asthma and wheezing in the first six years of life

Persistent airway hyperresponsiveness after neonatal viral bronchiolitis in rats

The association of viral and bacterial respiratory infections with exacerbations of wheezing in young asthmatic children

Viruses as precipitants of asthmatic attacks in children

Recurrent wheezy bronchitis and viral respiratory infections

Viruses as precipitants of asthma symptoms. I. Epidemiology

Risk factors for acute wheezing in infants and children: viruses, passive smoke, and IgE antibodies to inhalant allergens

Use of polymerase chain reaction for diagnosis of picornavirus infection in subjects with and without respiratory symptoms. / Clin Microbiol (\99?>)

Community study of role of viral infections in exacerbations of asthma in 9-11 year old children

Rhinovirus and influenza A infections as precipitants of asthma

Viral and bacterial infections in adults with chronic asthma

Association of viral and mycoplasma infections with exacerbations of asthma

Respiratory viruses and exacerbations of asthma in adults

Transtracheal aspiration studies in asthmatic patients in relapse with ^infective' asthma and in subjects without respiratory disease

Chronic sinus disease associated with reactive airway disease in children

Predictive value of respiratory syncytial virusspecific IgE response for recurrent wheezing following bronchioHtis

The relationship of RSV-specific immunoglobulin E antibody responses in infancy, recurrent wheezing, and pulmonary function at age 7-8 years

Exacerbations of asthma in adults during experimental rhinovirus infection

\ Immunohistochemical analysis of nasal biopsies during rhinovirus experimental colds

Rhinovirus upper respiratory infection increases airway reactivity in late asthmatic reactions

Respiratory infections: their role in airway responsiveness and pathogenesis of asthma

The effect of T cell depletion on enhanced basophil histamine release after in vitro incubation with live influenza virus

A common cold virus, rhinovirus 16, potentiates airway inflammation after segmental antigen bronchoprovocation in allergic subjects

Experimental rhinovirus 16 infection potentiates histamine release after antigen bronchoprovocation in allergic subjects

Lower airways inflammation during rhinovirus colds in normal and in asthmatic subjects

Nasal cytokines in common cold and allergic rhinitis

Glucocorticoids inhibit cytokine-mediated eosinophil survival

Rhinovirus enters but does not replicate inside monocytes and airway macrophages

A cell adhesion molecule, ICAM-1, is the major surface receptor for rhinovirus

Members of the low density lipoprotein receptor family mediate cell entry of a minor-group common cold virus

Interleukin-11: stimulation in vivo and in vitro by respiratory viruses and induction of airways hyperresponsiveness

\ Virus-specific CD8^ cells can switch to interleukin 5 production and induce airway eosinophilia

Eosinophil degranulation in the respiratory tract during naturally acquired respiratory syncytial virus infection

Activation of human eosinophils in vitro by respiratory syncytial virus

ACKNOWLEDGEMENTS Support for this chapter has come from grants NIH HL 44098, AI-26609, KO8-01828 and General Clinical Research Grant RR-03186.