| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Arizona Respiratory Center, University of Arizona, Tucson, Arizona
Correspondence and requests for reprints should be addressed to Fernando Martinez, M.D., Swift-McNear Professor of Pediatrics, Director, Arizona Respiratory Center, The University of Arizona, 1501 N. Campbell Avenue, Suite 2349 Tucson, AZ 85724-5030. E-mail: fernando{at}arc.arizona.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT ASSOCIATION BETWEEN VIRAL...HETEROGENEITY OF DETERMINANTS OF...CONCLUSIONSREFERENCES |
|---|
responses measured before any viral lower respiratory illness and increased interferon- responses measured at the time of the illness may predispose to such illnesses. Children in whom the former mechanism is involved should be expected to be more atopic later in life, whereas those with the latter mechanism should be less likely to be atopic. This may explain why early viral respiratory illnesses have been found to be both protective against and a risk factor for subsequent atopy in different studies. Current evidence thus suggests that different and often apparently contradictory mechanisms related to airway function, structure, and immune responsiveness may explain the association between viral lower airway illness in early life and subsequent asthma. Future preventive and therapeutic strategies will need to address the specific mechanisms that explain this association in different groups of subjects.
Key Words: respiratory syncytial virus • asthma • lung function • infant • lower respiratory illness
Viral infections are the most consistent factor associated with the expression of asthma, regardless of its phenotypic characteristics, of the age of the patient, and of the phase of the natural history during which they occur. Evidence of viral infection is present in almost all cases of acute airway obstruction occurring during the first years of life, at a time when it is still not possible to determine if the affected child will go on to develop asthma or will only have transient episodes of wheezing. Nevertheless, the preschool is the period of life with the highest incidence of asthma (that is, the rate at which new cases of asthma develop) (1), and therefore one in which the role of viral infection in asthma pathogenesis may be of the greatest importance. Viruses or their products are also detected in the majority of older children and adults during asthma exacerbations (2), suggesting that either viruses themselves or the immune response associated with the viral infection may play an important role in acute asthma flare-ups.
In spite of this clear, strong temporal association, the role of viral infections in the pathogenesis of asthma is not well understood. Three main hypothetical scenarios have been proposed. A first possibility, especially applicable to the pediatric age, is that viral infections may either alter the development of immune responses or interfere with the normal pattern of lung growth and/or regulation of airway tone. In this scenario, viruses would be causal agents in the inception of asthma; the obvious corollary of this hypothetical scenario being that control of the relevant viral infections (by decreasing exposure or by the availability of active or passive immunization) should decrease the incidence of asthma. A second scenario is that viruses are merely triggers of airway obstruction in subjects who have either a pre-existing alteration in airway function or structure or a susceptibility to develop immune responses that predispose to airway obstruction. In this second scenario, control of viral infection could decrease morbidity associated with asthma-like symptoms, but would not decrease the incidence of the disease.
The purpose of this review is to propose a third hypothetical framework, which could explain the apparently contradictory information available both from human studies and experimental models. In this view, the nature of the response to the different viruses associated with asthma will depend on the genetic background of the individual, on concomitant environmental exposures, and on the timing of the infectious episode, in relation to the degree of maturation of both the immune system and the airways. Therefore, there is no single pattern of responses and of potential long-term sequealae associated with the development of asthma-related viral infections.
In the next paragraphs the evidence that supports this paradigm will be reviewed. Special emphasis will be placed on the first years of life, because this age period offers the best available data to assess the potential role of viral infections in the inception asthma.
|
ASSOCIATION BETWEEN VIRAL INFECTIONS IN EARLY LIFE AND SUBSEQUENT ASTHMA |
|---|
|
TOPABSTRACTASSOCIATION BETWEEN VIRAL...HETEROGENEITY OF DETERMINANTS OF...CONCLUSIONSREFERENCES |
|---|
This conclusion is compatible with the results of the Tucson Children's Respiratory Study (3). In this prospective study, unselected children were enrolled at birth and followed up to adolescence and beyond. Most LRIs were ascertained by a physician during the first 3 years of life, and virologic studies were performed for each LRI. Given the state of diagnostic tools at the time the LRIs were studied (between 1980 and 1987), RV was not included among the viruses targeted for detection. RSV was found to be the most frequent virus isolated during LRI, especially in the first year of life, and children with RSV-LRI were found to be at significantly increased risk of subsequent wheezing during the school years, a risk that significantly decreased with age after age 6. The connection between RSV-LRI and subsequent asthma was independent of, and was not mediated by, atopy: children with RSV-LRI were not more likely to become skin test positive to local aeroallergens during the school years when compared with those who did not have LRIs. Interestingly, LRIs associated with wheezing occurring during the second and third year of life were much more likely to be associated with subsequent atopic asthma (11), and these "late" LRIs were less likely to be due to RSV and more likely to be associated with negative cultures than those occurring during the first year of life. The patterns emerging from the Finnish study and from the Tucson study are thus similar: first year of life wheezing LRIs are most often associated with RSV infection, whereas those occurring during the second and third year of life are increasingly less likely to be due to RSV, and more likely to be associated with RV infection or with negative cultures. LRIs occurring during the second and third year of life are more strongly associated with subsequent atopic asthma than those occurring in the first year of life, although RSV-LRI are associated with increased risk of subsequent wheezing, albeit not apparently related to atopy.
The results of the Finnish (9) and Tucson studies, however, diverge in certain aspects from those of a report by Sigurs and coworkers in Sweden (4, 12, 13). Sigurs and colleagues recruited children who were hospitalized for RSV-LRI as infants and followed them and a group of control subjects up to the age of 13 years. They reported that at ages 3.5, 7, and 13 years, children who had RSV-LRIs were significantly more likely to have reports of wheezing and a diagnosis of asthma than control children, but were also significantly more likely to be sensitized to local aeroallergens. The authors proposed the hypothesis that an increased risk of subsequent development of atopy could explain at least in part the association between RSV-LRI and subsequent asthma. This hypothesis is not compatible with the data from the Tucson study and the Finnish study (9), which suggested that the connection between RSV and subsequent wheezing was not mediated by an increased likelihood of developing atopic asthma. Interestingly, Juntti and coworkers (14) reported results from a study similar in design to that of Sigurs and colleagues, that is, a comparison of subsequent atopic status and prevalence of asthma in children hospitalized as infants with a confirmed history of RSV-LRI and control subjects. The authors also found an increased risk of subsequent asthma associated with RSV LRI, but reported that children who had RSV-LRI were less likely than control children to be atopic during the school years.
|
HETEROGENEITY OF DETERMINANTS OF LRI IN INFANCY |
|---|
|
TOPABSTRACTASSOCIATION BETWEEN VIRAL...HETEROGENEITY OF DETERMINANTS OF...CONCLUSIONSREFERENCES |
|---|
What follows is a description of four areas of research that have provided evidence for different mechanisms of acute LRI and for the potential association between early LRI and subsequent asthma. The objective is, more than to provide an exhaustive account of all the factors implicated in these processes, to illustrate the complexity of the processes involved. It is clear that, in each individual subject, overlap may occur between these different mechanisms, and that the interaction between them will determine probabilistically the degree of severity and the overall prognosis in each case.
Congenital Alterations of Airway Function/Size
There is now well established evidence that indices of airway size obtained from either partial maximal flow–volume curves or other techniques used to assess lung function in noncooperating infants are inversely associated with the risk of having wheezing LRI during the first years of life. The observation that infants with preexisting diminished maximal flows/higher airway resistance are more likely to present with airway obstruction during viral infections has been replicated in five separate studies (15–19). These observations are thus compatible with the hypothesis that diminished airway size is a risk factor for the development of wheezing illnesses during viral infections in early life. Moreover, it is also well established that the increased risk of wheezing that has been consistently reported for children whose mothers smoked during pregnancy may be mediated, to a significant extent, through the diminished maximal flows that are characteristically observed in children exposed to tobacco smoke products in utero (20). Given that airway resistance is a function of the fourth power of the radius of the bronchi, it has been proposed that the mechanism involved in this case is one of markedly increased functional obstruction in children with congenitally narrower airways (21). Moreover, because maximal flows increase with age relative to lung size (21), one would expect that this mechanism could mainly be active in the first years of life. Indeed, data from the Tucson Children's Respiratory Study (22) and from a similar longitudinal study in Perth, Australia (23) suggest that diminished airway function at birth is mainly associated with transient early wheezing, that is, with the development of one or more episodes of airway obstruction during the first three years of life, without further continuation of such episodes beyond that age. Interestingly, this is also the most frequently observed pattern of association between maternal smoking during pregnancy and childhood wheezing (24), suggesting once again that alterations in airway size due to both genetic or environmental factors may mediate this association.
The strong association between premorbid diminished airway function and transient earlier wheezing does not exclude the possibility that the former may also contribute to the risk for severe wheezing beyond the first 3 years of life. In fact, although relative to those in their peers, maximal expiratory flows improve during the preschool years in transient early wheezers (25), maximal flows are still significantly lower up to the age of 16 years in these children when compared to those of children who did not wheeze during viral infections in the first 3 years of life (unpublished observations). It is thus not surprising that, studying the children enrolled in the Perth study quoted earlier, Turner and coworkers observed that levels of lung function measured shortly after birth were significantly diminished in children who had reports of wheeze both at ages 6 and 11 years (26). Nevertheless, it is likely that congenitally diminished levels of lung function may not by themselves be sufficient to increase the risk for wheezing during the school years.
Congenital Alterations in the Regulation of Airway Tone
In some longitudinal studies, researchers were able to assess responses to either methacholine (27) or histamine (28) shortly after birth and before any LRI. These techniques are unfortunately cumbersome, and their continued use has also encountered significant resistance from ethics committees during the last 10 years. As a consequence, the data available are much scantier and less conclusive than that for earlier postnatal airway function tests without provocation, as described in the previous point. Nevertheless, some issues appear to be well established. Bronchial responses to histamine or methacholine are not associated with the level of baseline lung function in infants (29), and therefore, these two measurements assess potentially different aspects of airway physiology. Clarke and colleagues reported a significant association between bronchial hyperresponsiveness (BHR) to histamine and the subsequent risk of wheezing, presumably in relation to viral infections (30). The effects were stronger in girls than in boys. Among children enrolled in the same study from Perth quoted earlier, a group of slightly less than 100 were tested for BHR using histamine both shortly after birth and at the age of 6 years (29). The authors reported that early postnatal BHR to histamine was unrelated to BHR measured in these same children at age 6. However, early postnatal BHR was associated with wheezing at age 6 and with the level of lung function measured at that same age. Moreover, BHR measured at age 6 contributed to the risk of wheezing at that age independently of early postnatal BHR. BHR was not assessed in the early postnatal period in the Tucson Children's Respiratory Study, but BHR to cold dry air was measured at age 6 and BHR to methacholine and responses to albuterol were separately measured at age 11 (3, 31). Much like in the Perth study, BHR at age 6 was associated with wheezing at that age but was unrelated with a history of LRI in early life. However, although both responses to albuterol and methacholine were associated with current wheezing at age 11, only the former was associated with a history of RSV-LRI in the first 3 years of life.
The interpretation of this complex set of data is not straightforward. Most likely, both bronchial provocation tests performed shortly after birth and responses to albuterol measured during the school years detect intrinsic characteristics of the airways that are unrelated to airway inflammation. This is supported by the reported association of both these measures with two known polymorphisms in the coding region of the ß2-adrenergic receptor gene (32, 33) and by the lack of association of either of these measures with markers of atopy such as skin test reactivity. Therefore, it is possible that a congenital (or "intrinsic") component of the regulation of airway tone may predispose children to wheezing during viral infection in early life and to continue wheezing during the school years, thus explaining the connection between these two phenomena, independent of atopy. As describe earlier, however, the strength of this association decreases with age between 6 and 13 years (3), and the factors that determine this decreasing trend have not been elucidated. Interestingly, data from a longitudinal study in Dunedin, New Zealand, suggest that BHR to methacholine tends decreases between ages 9 and 15 years, but that this trend is only observed among nonatopic children, whereas the level of BHR tends to persist among atopic children (34). Therefore, a potential explanation for the diminishing trends for the association between RSV-LRI and subsequent wheezing could be the decreasing "intrinsic" BHR occurring during the school years.
Altered Immune Responses to Viral Infection
Several lines of research suggest that infants may be more prone to develop lower airway involvement during viral infection because they have immature immune responses. For example, infants produce low levels of protective antibodies against RSV (35), and cytokine responses of both the T helper-1 (Th1) and Th2 types to nonspecific stimuli are weaker in infants than in adults, with deficits in the former being much more prominent than those in Th2-like cytokines (36, 37). Moreover, epidemiologic data suggest that interferon- (IFN-) responses by peripheral blood mononuclear cells measured during the first months of life strongly predict the subsequent development of wheezing illnesses, presumably due to viruses, during the first year of life (38). A study of children hospitalized with severe RSV seemed to indirectly support this finding: children requiring mechanical ventilation had significantly lower IFN- levels in nasopharyngeal secretions than those not requiring ventilatory support (39). However, one recent study found that levels of IFN- were elevated in nasal secretions of infants with RSV infection, and that they correlated positively with the likelihood of wheezing during the infection (40). IFN- was also found to be correlated with leukotriene levels in nasal secretions, and the latter has been found to be correlated with severity of acute bronchiolitis (41). These contradictory findings are surprising and difficult to explain. It is possible, however, that these studies may have targeted different groups of children, in whom different mechanisms are predominantly at work. For example, low baseline IFN- production may be a risk factor for subsequent wheezing in infants in whom the mechanism of disease is related to increased susceptibility to viral infection due to a delayed development of IFN- responses. Interestingly, infants with delayed IFN- responses are also more likely to become sensitized to aeroallergens later in life (42). Therefore, delayed IFN- responses could explain the link observed in certain studies between RSV illness in early life and subsequent allergic sensitization, particularly if more children belonging to this at risk category were enrolled in those studies (5). On the other hand, high producers IFN- could be a different susceptibility group, in whom the pathogenesis of the acute RSV illness could be due to excessive IFN- responses, which cause an immune-mediated disease mechanism. Paradoxically, these children should be expected to be less atopic later in life, a finding that is compatible with the report by Juntti and coworkers (14) cited earlier. In studies from random samples of the population in which both these types of subjects are enrolled (3), the expected result would be no association between RSV illness and atopic sensitization. Unfortunately, no studies are available in which IFN- responses were measured before and during RSV illness in the same children, and therefore, the confirmation of the existence of these two different, IFN-–related mechanisms for RSV illness remains speculative. However, studies in which an attempt was made to identify children who would or would not go on to have persistent wheeze and/or atopy after an acute LRI in early life support the hypothesis that the latter is not a homogenous group. Martinez and colleagues showed children with wheezing LRI who subsequently developed asthma had significantly higher total serum IgE and eosinophil levels during the acute phase than during the convalescent phase of the LRI, whereas those who did not go on to develop asthma showed no such acute responses (43). Moreover, children who had nonwheezing LRI were less likely to have high serum IgE levels later in life than those who did not have LRI (44). The data thus suggests that outcome of early life LRI may vary according to the nature of the immune response to the virus and that, depending on which group of subjects predominates in each study, early LRI may be found to be associated with increased or decreased likelihood of subsequent atopy.
Immune mechanisms other than those associated with IFN- responses have also been implicated in the association between RSV illness and subsequent wheezing. Bont and colleagues (45) reported that increased monocyte interleukin-10 responses during acute RSV illness were directly associated with the likelihood of subsequent persistent wheezing, and this effect was independent of IFN- responses. It is this likely that a number of different immune mechanisms may be involved in the connection between RSV illness in early life and subsequent persistent wheezing and asthma.
Genetic Determinants of RSV Illness
In spite of the extraordinary advances in the development of new technologies to study the genetics of complex disease, there have been surprisingly few studies of the genetics of early life LRI and its potential association with the subsequent development of asthma. This paucity may be in part due to the intrinsic difficulty to demonstrate that familial aggregation of LRI is attributable to common genetic background and not to common environment. For example, if young siblings were found to be more susceptible to RSV illness if their living siblings also had RSV, the most reasonable explanation would be a common (or mutual) infectious source. Moreover, if early life LRI is a highly heterogeneous condition and, as suggested earlier, both low and high responses of the same cytokine (e.g., IFN-) are associated with increased risk of acute illness, studies that included all affected children as probands could not be very successful.
Hull and coworkers first reported that the A allele of a single nucleotide polymorphism (SNP) at position –251 from the transcription start site in the promoter region of IL-8 was associated with severe RSV illness requiring hospitalization (46). This group of researchers also reported that the same allele (and other polymorphisms linked with it in the same haplotype) was associated with the likelihood of continued wheezing 5 to 6 years after an episode of RSV requiring hospitalization (47). Functional studies indicated that the haplotype linked with the A allele was associated with increased transcription rates for interleukin-8 (48), suggesting that increased interleukin-8 responses could be risk factors for persistent wheezing after RSV. However, Heinzmann and colleagues (49) studied the same SNP in older children, and found that susceptibility to asthma was associated with the T allele, exactly the opposite effect to that reported by Hull and coworkers. If these apparently contradictory findings are replicated in other populations, they would confirm that the mechanisms that determine the association between early life events and subsequent asthma are highly heterogeneous. It is possible that, much like seems to be the case for IFN-, inflammatory mediators such as interleukin-8 may have opposite effects on disease susceptibility, depending on the context in which they are expressed.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTASSOCIATION BETWEEN VIRAL...HETEROGENEITY OF DETERMINANTS OF...CONCLUSIONSREFERENCES |
|---|
|
FOOTNOTES |
|---|
(Received in original form April 22, 2005; accepted in final form June 21, 2005)
|
REFERENCES |
|---|
|
TOPABSTRACTASSOCIATION BETWEEN VIRAL...HETEROGENEITY OF DETERMINANTS OF...CONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. L. Collins and B. S. Graham Viral and Host Factors in Human Respiratory Syncytial Virus Pathogenesis J. Virol., March 1, 2008; 82(5): 2040 - 2055. [Full Text] [PDF]
|
|
|
|
 |
|
 M. U. Ferreira, G. Rubinsky-Elefant, T. G. de Castro, E. H. E. Hoffmann, M. da Silva-Nunes, M. A. Cardoso, and P. T. Muniz Bottle Feeding and Exposure to Toxocara as Risk Factors for Wheezing Illness among Under-five Amazonian Children: A Population-based Cross-sectional Study J Trop Pediatr, April 1, 2007; 53(2): 119 - 124. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
