Temporal Variation

Following a well-documented epidemic of asthma deaths reported in children and young adults (5–34 years) in several, but not all, industrialized HICs in the 1960s, Burr reported an increase in asthma hospital discharges of 5- to 14-year-old children in Wales. This prompted a series of cross-sectional surveys at intervals, which demonstrated a doubling in the prevalence of ever having asthma in children in South Wales from 6% in 1973 to 12% in 1988; 15 years later. A similar trend was seen in 6–11-year-old children in the US National Health and Nutrition Examination Survey (NHANES) between the early and late 1970s, and in a number of other countries over a comparable epoch from the early 1970s through the 1990s, including the UK, Scandinavia, Israel, New Zealand, and Australia. These mainly relied on parental report of wheeze and/or asthma in repeat cross-sectional surveys of parents in the same geographical setting. By 2000, there were indications that the rising prevalence in childhood asthma in high-income, developed countries was coming to an end. A Swiss study based on national surveillance data suggested that this reversal of the previous trend could be explained by fewer consultations for allergic asthma. One of the longest running surveys of asthma and allergy symptoms started in Aberdeen, UK, in 1964, and has used the same methods in repeat studies of the same age group of children in an economically stable environment with low levels of migration in studies spanning a 50-year period. The original surveys showed contemporaneous rises in the prevalence of asthma, eczema, and rhinitis in this age group of children, pointing to a common etiology mediated through influences on allergic mechanisms. However, later studies, which confirmed a “flattening” of the previous trend for increased prevalence of these conditions similar in magnitude and direction to results from comparable HICs, showed a continued rise in the lifetime prevalence of eczema and hay fever, thus throwing doubt on a common etiology of these conditions. Therefore, the summation of evidence, largely from repeat surveys using identical methodology in different settings, suggests that reported asthma prevalence and the prevalence of asthma symptoms showed an increase from the early 1990s until around the end of the 20th century. However, there are a number of considerations that need to be taken into account before accepting this as evidence for a true increase in disease prevalence. These include changes over time in diagnostic coding, such as changes in the International Classification of Diseases (ICD) definitions; diagnostic preference, such that wheezing illnesses may or may not be labeled as asthma; increased population awareness leading to more presentations to health care with asthmalike symptoms; and systematic bias in questionnaire responses. Although repeat surveys were carried out in comparable groups from similar strata of society, they were not the same individuals or families, so they may have differed in their responses without there being a true change in symptoms. Subsequent analyses have suggested a diagnostic shift explaining part of the observed increase in prevalence, but there is still evidence that at least part of the rise in asthma prevalence was real. The ISAAC study has provided a global perspective of these changes in symptoms over time through incorporation of a repeat survey in the third phase of the study (ISAAC phase III), which was designed to describe time trends and factors associated with these in the centers that completed phase I, which established baseline estimates and their geographical variation. Phase III followed the initial phase by at least 5 years with an average interval of 7 years. The study gathered data from over 190,000 6- to 7-year-olds in 37 countries and over 300,000 13- to 14-year-olds in 56 countries. These showed that between 1992 and 1998, and 1999 and 2004, centers were about equally distributed between substantial increases and decreases in the prevalence of asthma symptoms, whereas in 6- to 7-year-olds more centers showed an increase than a decrease in prevalence ( Fig. 42.2 ). Analysis of the relationship of time trends between asthma, eczema, and rhinitis showed that the strongest correlation was between eczema and rhinoconjunctivitis, and the weakest between eczema and asthma. However, the relationships overall showed consistency across the study period, suggesting that similar factors are affecting them at a global level, and there was no strong association with gross national income across the low-income countries to HICs comprising the study centers, despite differences in the proportion of atopic and nonatopic asthma between high- and low-to-middle income countries; nonatopic asthma is more prevalent in the latter.

World map showing the direction of change in the prevalence of asthma symptoms for the 6- to 7-year age group (A) and 13- to 14-year age group (B). Each symbol represents a center. Blue triangle = prevalence reduced by ≥1 Standard Error per year. Green square = little change (<1 Standard Error). Red triangle = prevalence increased by ≥ 1 Standard Error per year.

(From Asher MI, Montefort S, Björkstén B; ISAAC Phase Three Study Group. Worldwide time trends in the prevalence of symptoms of asthma, allergic rhinoconjunctivitis, and eczema in childhood: ISAAC Phases One and Three repeat multicountry cross-sectional surveys. Lancet . 2006;368(9537):733-743.)

To address the question of whether reported increases in the prevalence of reported symptoms of asthma represent a true underlying change in disease incidence and/or natural history, it is instructive to consider other markers of disease activity that may indicate whether the figures reflect a change across all grades of asthma severity or the inclusion of a greater proportion of milder disease; the latter could represent diagnostic shift to include milder cases or a true change in asthma severity or its management. For example, better drug treatments or behavioral shifts leading to greater adherence to asthma medication, could impact health care utilization without affecting the underlying disease prevalence. Similarly, changes in health care organization over time, such as greater emphasis on emergency room care, could reduce hospital admission rates without changes in asthma attack rates, perhaps due to direct discharge of more moderate cases. Surveillance data from the United States has confirmed a 4.6% annual increase in asthma 12-month prevalence in children (<17 years) from 1980 through 1995 with a stabilization of current asthma prevalence since 1997. Asthma-related emergency department visits remained relatively stable between 1992–2006 while asthma deaths in US children increased by 3.2% per year from1980 through 1996 and have since decreased. These data are broadly supported by similar trends in the UK ; children (<14 years) had increased reported prevalence of asthma symptoms and about an eight-fold increase in primary care consultations for asthma between the 1970s and mid-1990s. This was accompanied by increased hospital admissions for asthma until 1990, since when they declined ( Fig. 42.3 ). The evidence of increased prevalence of asthma of all grades of severity during the last few decades of the 20th century support a true increase in disease prevalence rather than a broadening of disease classification.

Hospital admissions for asthma by age, England and Wales 1958–2003.

(From Anderson HR, Gupta R, Strachan DP, Limb ES. 50 years of asthma: UK trends from 1955 to 2004. Thorax 2007 Jan;62(1):85–90.)

Spatial Variation

The potential of investigating spatial or geographical variations in asthma was exemplified by the landmark studies in Germany following its reunification in 1990. The distinct advantages of this setting were the shared genetic ancestry of the populations of East and West Germany coupled with the contrasting lifestyles and environments of the German Democratic Republic in the East and the Federal Republic of Germany in the West. Thus, it was possible to investigate different environmental influences in two genetically similar groups. At the time, the contemporaneous increase of asthma prevalence in the West and the marked increases in road traffic densities had indicated traffic-related pollution as a possible contributor to asthma etiology. It was reported that the prevalence of atopic sensitization, hay fever, and asthma was higher in children born in the West, where traffic-related pollution dominated, whereas bronchitis prevalence was higher in the East, where industrial sources of pollution persisted. The first phase of the ISAAC study mapped the prevalence of asthma in 6- to 7- and 13- to 14-year-old children using comparable survey methodology, and showed the highest prevalence in countries where English was the first language and which could be described broadly as having a “Westernized” lifestyle, and in Latin America. However, some of the regions with the highest reported levels of air pollution had generally low asthma prevalence, pointing away from air pollution as a major cause of asthma onset (as opposed to exacerbations). The follow-up phase 3 ISAAC report continued to show striking variation in asthma prevalence ( http://isaac.auckland.ac.nz/story/methods/maps.php#mapMenu ) between countries with similar regional patterns to the phase 1 study published 11 years earlier, and was mirrored by the prevalence of symptoms of severe asthma; high prevalence was associated with HICs. Intercountry comparisons of asthma prevalence present an opportunity for ecological analyses of the differences in lifestyles and exposures associated with asthma in these settings. Although such approaches can give clues about possible etiologic factors, population genetic differences could influence the environmental exposures that are instrumental in asthma onset in different populations. This is particularly important in the study of complex, polygenic diseases, such as asthma, that arise from interactions between genes and environment. However, because the confounders of associations between risk factors and asthma almost certainly differ between regions, an opportunity is presented to strengthen causal inference of exposure-outcome links. If associations between a putative risk factor and asthma are replicated in settings with markedly different lifestyles and environments, it is likely that confounding plays a lesser role in explaining such relationships, making the risk factor the more probable (causal) explanation.

One key difference between countries with contrasting asthma prevalence in the ISAAC studies is the degree of urbanization. An attempt has been made to address this question using satellite images to estimate the extent of urban development and matching these to ISAAC study centers. Despite some, but not all, regional studies, mostly set in low-income countries, reporting links between increased urbanization and asthma prevalence, this global perspective showed only weak evidence of any association between urban development and asthma prevalence, including severe asthma symptoms, in children. Considering asthma prevalence in HICs, there is clear evidence from the United States that asthma prevalence is higher in urban areas with a high proportion of ethnic minority populations, and this is reflected in higher indicators of morbidity and mortality in children of Hispanic and African-American origin. Many possible explanatory factors for this observation have been suggested, including genetic predisposition and environmental exposures associated with social disadvantage, such as outdoor and indoor air pollution, allergens, infectious agents, and diet. These will be considered further under etiologic factors associated with asthma. A further ecologic association arising from the ISAAC studies is the low prevalence of reported asthma in countries where a high proportion of the population lives in rural areas with reliance on subsistence farming. These observations are consonant with studies from rural communities, mainly in middle Europe, but replicated in other settings, where the prevalence of asthma and allergy has been reported to be lower in children growing up on farms and exposed to livestock and other factors, such as drinking unpasteurized milk. The common mediating factor of this protective effect is now believed to be early life exposure to microorganisms and their cellular constituents, including endotoxins. Higher levels of endotoxin in dust collected from farms has been associated with lower prevalence of sensitization and allergic diseases, although later exposure from household dust can exacerbate asthma symptoms in urban settings. Recent evidence suggests that the protective effect of endotoxin exposure in early life is mediated through induction of an enzyme (A20) in the pulmonary epithelium that modifies interactions between the epithelium, antigen-presenting cells, and the innate immune system.

Of particular interest in the geophysical variation of asthma prevalence is the effect of migration between settings with widely varying rates of asthma. Migrants from countries with low asthma prevalence to high prevalence countries have acquired the rates of asthma in the local population, with the effect dependent on the time spent in the adopted country. There is also evidence that the younger age at which migration occurred in these circumstances, the higher the prevalence of later asthma, lending support to the early origins hypothesis that environment in infancy and early childhood is critical for the development of diseases through the life course. The ISAAC study group has recently analyzed data on migration and reported that being born outside the country of study was associated with a lower prevalence of asthma in both 6- to 7- and 13- to 14-year-old children. This protective effect was restricted to affluent countries and was not evident in low-income settings with low asthma prevalence. It was also lost with increasing time spent in the host country, and points to environmental factors as opposed to genetic factors being important in the variation of the prevalence of asthma between countries.

Variation by Gender and Ethnicity

Asthma prevalence is disproportionately higher in black and ethnic minority populations than in white children of European ancestry. The explanations for such disparities are likely to be multifactorial and include genetic and environmental influences, urban living, poverty, and access to health care. Although asthma has high heritability, racial and ethnic minorities are underrepresented in genetic studies of asthma (see: http://www.ebi.ac.uk/gwas/home for a catalog of genome wide association studies [GWAS]) and the current commercially available chips for genome wide, SNP screens were derived from populations of mainly European descent. In Southern California, a study using genetic markers of ancestry in Hispanic children showed that those with a lower proportion of African ancestry were at a lower risk for asthma than those who had major contributions from European or Amerindian ancestries. In common with many other developed, Anglophone countries, asthma prevalence in children in the United States approximately doubled from the early 1980s until the late 1990s. At the beginning of this period, there was little disparity between white and black children, whereas by 2010, the prevalence in African-American children was twice that of their white peers. During the period from 2000 onwards, when the previous increase in asthma prevalence in US children has plateaued, the disparity between asthma in black and white children also stopped increasing. This was largely due to a slowdown in the increasing prevalence of non-Hispanic black children, although asthma prevalence in some groups continued to rise, including those from a deprived socioeconomic background. In addition to higher asthma prevalence, African-American and Hispanic children in the United States were reported previously to have increased rates of hospitalization, emergency department use, and mortality due to asthma than their white counterparts, although a national analysis of in-patient data has reported increased mortality only for Native American children and not for other racial or ethnic groups. In the same study, the length of hospital stay after admission for an asthma attack was related to indicators of social deprivation but was shorter in Asian than in white children. In the UK Leicester Respiratory cohort, which has a high proportion of South Asian participants, a higher prevalence of ever having wheezed was reported in Asian compared with white children, but current symptoms of wheeze and night cough at age 6 years were similar between the two groups. In this study, associations with risk factors were also similar between white and South Asian children. Others have reported racial disparities in some associations, including body mass index (BMI), which was more strongly associated with asthma in US children who were male, non-Hispanic black, and living in deprived neighborhoods, traffic-related air pollution (TRAP), which was associated with hospital readmission in white but not African-American children, and possibly airway inflammatory responses to environmental tobacco smoke (ETS) in adolescents. Intriguingly, racial differences in treatment responses in severe, therapy-resistant asthma have been reported, with poorer fractional exhaled nitric oxide (FeNO) responses in black than white children to intramuscular steroid injections. The mechanism of this has yet to be discovered, although differences in genes affecting pharmacologic responses are likely to be important and could pave the way to personalized approaches to treatment.

Asthma is more common in males in early childhood but a gender switch in asthma prevalence occurs around the time of puberty, so asthma prevalence in young adults is higher in females than males; although reporting bias in favor of asthma in young males has been suggested to explain at least part of the apparent gender switch. The ISAAC study reported a higher prevalence of asthma in males at 6–7 years and a higher prevalence in females at 13–14 years with considerable variation between centers. In the UK longitudinal Isle of Wight cohort, males had a higher prevalence of asthma up to 10 years, which had reversed to higher female prevalence by 18 years, with more wheezing females gaining an asthma diagnosis between 10 and 18 years. Some cross-sectional studies have suggested that the sex discordance in asthma observed through the peripubertal period has become less exaggerated in recent years. There is some uncertainty about the precise age at which the gender switch occurs and its relation to the onset and timing of puberty. Symptom severity has been reported to be greater in boys in early childhood, transitioning through 7–9 years to be greater in females from 10 to 17 years, with a marked influence of puberty on symptom progression. Early menarche in girls has been associated with an increased risk of incident asthma. These observations have implicated estrogenic effects as potential mediators of the increased prevalence and severity of asthma in females around the time of puberty. Estrogen receptors are expressed on many cell types, including airway smooth muscle. Their expression in immune cell lines is associated with enhancement of allergic inflammation through Th2 skewing of immune responses, isotype switching of B cells to immunoglobulin E (IgE) production and mast cell degranulation. On the other hand, some evidence points away from estrogenic effects, such as fluctuations in bronchial responsiveness through the menstrual cycle being associated with alterations in serum testosterone concentrations, and it is probable that multiple factors, including sex differences in biological and lifestyle variables, are associated with the changes in asthma prevalence that occur during transition to early adulthood.

Outcomes of Childhood Asthma

A high proportion of asthma presents with symptoms in early childhood, when recurring, intermittent wheeze is very common, affecting 30%–50% of infants and young children. A substantial number of children with early onset wheezing will have symptom remission during childhood, especially those with milder symptoms and an absence of severe attacks requiring hospital admission. Prospective longitudinal studies of asthma have been of vital importance to understanding the outcomes of asthma in childhood. These avoid the inevitable pitfalls of recall bias whereby adults with asthma are more likely to report symptoms in childhood than those whose asthma has resolved. The Melbourne Children’s Asthma Study is one of the longest-running studies of asthma in childhood, having recruited 7-year-old children with asthma and wheeze in 1964, with a supplemental recruitment of those with more severe disease at the age of 10 years. Outcomes from this seminal study have now been reported through 50 years of age and have shown that a number of factors in childhood; asthma severity, female sex, and coexistent hay fever were associated with asthma presence in mid-adulthood. Furthermore, severe asthma and asthma in childhood, compared with wheezing bronchitis and healthy controls, were associated with lower forced expiratory volume in 1 second (FEV ₁ ) and the ratio of FEV ₁ to forced vital capacity (FEV ₁ /FVC) in adults; severe asthma was strongly associated with COPD despite a high proportion of this cohort being never smokers. The lung function differences seen at age 50 years were not explained by between-group differences in the rate of FEV ₁ decline in adulthood, suggesting that these differences were established in childhood and persisted (“tracked”) to adult life. This observation is consistent with results of a large comparative study of adult cohorts, which showed that low FEV ₁ in early adulthood followed by physiological rates of decline was associated with COPD in a substantial proportion of participants, and supports accumulating evidence about the early life origins of COPD. Another long-running cohort study of asthma in children from Dunedin, New Zealand, also showed evidence of low lung function in adults with persistent wheezing from childhood and had previously reported allergic sensitization, airway hyperresponsiveness, female sex, and smoking as risk factors for the persistence of symptoms in early adulthood and showed similar tracking of low lung function from recruitment at age 9 years. In the UK, a follow-up at 50 years of children recruited in 1964 confirmed associations of childhood asthma with COPD in the absence of accelerated decline of FEV ₁ during adulthood. A report from the Childhood Asthma Management Program (CAMP) research group categorized children on the basis of trajectories of repeat lung function measurements and showed that children with persistent asthma with reduced lung function growth were at risk for fixed airflow obstruction and possible COPD in early adulthood. This subject is discussed in more detail in Chapter 15 .

Hospitalization

As asthma prevalence increased in children in the early 1980s, there was a parallel increase observed in asthma hospitalization rates in the same countries, suggesting that the increased prevalence was not wholly explained by the increased recognition of milder cases. However, since the 1990s, hospital admission rates for asthma in children stabilized or decreased in many developed countries. In a US survey, the rates of one or more asthma attacks in the past 12 months in children remained stable over the period 2001–2010. However, these trends were not replicated in all settings. A recent Scandinavian study analyzed hospital discharge codes over a 35-year period and concluded that the asthma admission rate for children had remained stable in the time period 1977–2012 at around 1 per year per 1000 children at risk. The rate of admission for asthma in children in Spain fell from 20.5 to 18.8 per 100,000 from 2002 to 2012, while during an almost contemporaneous period in neighboring France (2002–2010) the age standardized rate of admission in children increased by 2.5% per year. Therefore, variations in time trends of hospital admission rates for asthma likely depend on factors other than the background prevalence of severity. Neighborhood deprivation was associated with increased odds for hospitalization even after taking account of maternal socioeconomic indicators in a Swedish study that included over 17,500 children admitted to hospital with asthma. In a US study of geographical access to care in two states, severe asthma outcomes including emergency department visits and hospitalization differed in their associations with access to care according to a number of factors, including which outcome measure was used.

Many factors have been identified that are associated with increased risk of hospitalization for asthma in children. A recent systematic review of ETS exposure concluded that children with asthma exposed to tobacco smoke were nearly twice as likely to be hospitalized for asthma. The introduction of smoke-free legislation in Scotland in 2006, prior to which asthma admissions were rising at around 5% per year, was associated with a decline of 18.2% per year in both preschool and school-aged children and was independent of socioeconomic status. Similar findings were reported from England when national legislation was adopted in 2007.

Poor adherence to treatment has been identified as a risk for severe asthma outcomes and may be manifested by increased reliance on reliever medications. A US study used a controller-to-total asthma medication indicator in children and found that a ratio of less than 0.5 was associated with increased risk for subsequent hospitalization or emergency department attendance. Psychological stress induced by the death of a close relative was associated with increased risk of asthma hospitalization in adolescents. Acute exposures to environmental factors can trigger transient increases in hospitalizations for asthma. These include extremes of weather conditions, seasonal factors, including allergen exposure and respiratory infections, and air pollution. Genetic factors may also increase the risk of severe asthma exacerbations in relation to specific exposures. A polymorphism of the β-2 receptor has been associated with an increased risk of hospitalization, emergency care, or intensification of treatment in children prescribed long-acting sympathomimetic bronchodilators, and genetic variants in endotoxin signaling pathways can interact with indoor endotoxin exposure to increase the risk of hospitalization in children with asthma.

Severe Asthma

Although the majority of children with asthma have mild-to-moderate disease that can be managed effectively with appropriate controller medications, a proportion have severe and/or therapy-resistant disease. International consensus guidelines on the definition, evaluation, and management of severe asthma in children have been published and its heterogeneity has been documented using cluster analysis in a US study. A recent European study has used a case-control design to identify features of severe asthma in children compared with those with persistent but nonsevere disease. In a multivariate analysis, severe asthma was characterized by sensitization to food allergens (previously reported as a risk factor for life-threatening asthma), hospitalization and emergency visits for asthma, symptoms in response to physical activity, and lower lung function, although half the cases had FEV ₁ within the normal range. Interestingly, the cases could not be distinguished from controls by the home environment (apart from tobacco smoke exposure), parental education, or adherence to treatment.

Asthma Deaths

An epidemic of reported asthma deaths was observed in some, but not all, industrialized HICs in the latter half of the 20th century, beginning in the 1960s. Registered causes of death of 5- to 34-year-olds were analyzed from 1959 to 1979 and found to increase in New Zealand, Australia, England, and Wales, but not in the United States, Canada, or West Germany. As with all temporal trends, it is important to consider changes over time in disease classification, the way death certificates are completed and recorded, and a shift in diagnostic preferences. Of note, the return to preepidemic death rates was slower in New Zealand than in other countries, and New Zealand experienced a “second epidemic” in the 1970s, suggesting specific geographical factors in asthma management. This led to speculation that inhaled sympathomimetic drugs were implicated, supported by ecological evidence of higher sales of these (and other) asthma drugs in New Zealand, and some evidence that regular corticosteroids were underused. Further evidence from case control studies implicated the potent β-sympathomimetic drug, fenoterol in deaths due to asthma. A recent Cochrane review of another β-sympathomimetic drug implicated in asthma deaths, formoterol, considered 20 studies in adults and 7 studies in children and adolescents. All deaths occurred in adults, and the authors concluded, on the basis of low-quality evidence, that there was weak evidence of increased adverse events associated with regular use of formoterol in children However, two recent multicenter RCTs comparing combination corticosteroids/long-acting bronchodilators in adults and adolescents (>12 years) using budesonide/formoterol and in children (4–11 years) using fluticasone/salmeterol showed noninferiority of combination treatment compared with inhaled corticosteroid alone in relation to serious asthma-related adverse events. Deaths from asthma in childhood are relatively uncommon in HICs, having declined since the 1990s. National surveillance surveys have been used to attempt to identify factors associated with increased risk of asthma mortality. In New South Wales, Australia, 20 children died from asthma over the 10-year period, 2004–2013 ( http://www.ombo.nsw.gov.au/news-and-publications/publications/annual-reports/nsw-child-death-review/nsw-child-death-review-team-annual-report-2013 ). Risk factors for death included older age, low socioeconomic status, psychosocial problems, and Asian or Pacific Island racial background. A high proportion of those who died had a record of hospitalization in the previous 12 months; other factors identified were lack of follow-up care after admission, poor adherence to medication, and suboptimal asthma control. The Royal College of Physicians in the UK led a national confidential enquiry into all asthma deaths occurring in the year commencing February 2012 ( https://www.rcplondon.ac.uk/projects/national-review-asthma-deaths ). Of 195 people that died, the majority had asthma diagnosed as adults and only 28 deaths were in people younger than 19 years. Of all those who died, important factors identified were previous hospital admissions, attendance at an emergency department in the 12 months prior to death, failure to seek medical assistance during the final attack, and lack of specialist follow-up care. A low proportion of patients had personal asthma action plans or had received an asthma review in primary care in the year before death. Avoidable factors were identified in almost two-thirds of asthma deaths, including smoking and exposure to tobacco smoke, poor adherence to medications, psychosocial problems, and nonattendance at review appointments. In children and young people, poor recognition of the risk of adverse outcome was found to be of particular relevance.

Temporal Progression of Symptoms

In a seminal paper from the Tucson Children’s Respiratory Study based on wheezing history from birth to 6 years, three discrete phenotypes were described with varying associations with lung function, allergic sensitization, and outcome. This study was instrumental in establishing the concept of transient early wheezing in the first few years after birth as a condition distinct from asthma and most likely to arise as a consequence of an airway developmental disorder with low airway function measured soon after birth, a lack of association with allergic sensitization, and loss of symptoms by 6 years of age. Subsequently, in recognition of the phenotypic heterogeneity of asthma and wheezing illnesses in children, several groups have taken various approaches to classifying wheezing from early childhood by temporal progression of symptoms. Most have used statistical approaches that involve clustering symptom patterns using either a single cardinal symptom (wheeze) or several symptoms combined. These methods are generally regarded as data driven with no prior stipulations about the number or characteristics of the phenotypes derived. In a large, population-based study in the UK, latent class analysis was used to derive five different temporal patterns of wheezing from birth to 7 years. The important point to note here is that latent classes are not observable, so individual subjects cannot be assigned to a specific phenotype, but instead have a set of probabilities of belonging to one or several classes. Latent structures in the data are simply a way of describing variations that may have distinct underlying biological pathways (endotypes) with discoverable and preventable risk factors. Therefore, these methods are not directly translatable to the clinic but can be regarded as hypothesis-generating approaches. Replication studies in several different cohorts using latent class analysis or variants of this method confirmed the general description of these temporal classes or phenotypes. Wheezing phenotypes that started early (within 6–18 months after birth) and persisted through mid-childhood were particularly associated with a family history of asthma, evidence of allergic sensitization, and reduced airway function in later childhood. Similar associations with lung function outcomes have been reported in a Scandinavian cohort using simple categorization to different wheezing patterns; early persistent wheeze was associated with low FEV ₁ at age 16 years and with evidence of small airway dysfunction. Although all wheezing phenotypes compared with children who never wheezed show some decrement in measures of airway function in later childhood, persistent wheeze has also been associated with low growth of FEV ₁ through adolescence. Extended latent class analysis of wheeze to age 16 years has recently been reported in a UK birth cohort with similar outcomes associated with persistent wheezing phenotypes; increased bronchodilator responsiveness, lower FEV ₁ /FVC, and higher FeNO levels ( Fig. 42.4 ). Using a similar statistical approach, but including multiple data dimensions in the Leicester cohorts in the UK enabled identification of three wheezing phenotypes and two of chronic cough. These have subsequently been replicated in an independent, population-based cohort with two distinct phenotypes showing consistency between the discovery and replication samples; atopic persistent wheeze and transient viral wheeze. The overall synopsis of this series of studies is that wheeze that begins in early childhood and becomes persistent is associated with a poorer prognosis than other wheezing patterns. However, the original premise to identify factors influencing different phenotypes has not materialized; known associations of transient early wheezing and low airway function in early childhood with exposure to tobacco smoke during pregnancy and factors associated with a lower incidence of lower respiratory infections in early childhood have emerged from association studies with derived phenotypes, but there were few exposures that showed clear differentiation between different patterns of wheeze. The derived phenotypes have been shown to have external validity in a German comparison of latent class-derived and clinical phenotypes and comparison between these two enabled the discovery of a previously unrecognized clinical phenotype of intrauterine tobacco smoke exposure, decreased lung function, increased genetic risk for asthma, but without an established asthma diagnosis or treatment associated with unremitting wheeze. A joint modeling approach has also been applied to data in a cohort from Manchester, where the availability of links to clinical records allowed latent class analysis of parent reported and physician-confirmed wheezing. This enabled the separation of children with persistent wheezing into those with mild, controlled disease and those with “persistent troublesome wheeze” with the latter displaying reduced lung function, increased airway responsiveness, and a marked increase in exacerbations and hospitalizations compared to the other classes.

Estimated prevalence of wheezing at each time point from birth to years for each of the six wheezing phenotypes identified by using latent class analysis in 12,303 children.

(From Granell R, Henderson AJ, Sterne JA. Associations of wheezing phenotypes with late asthma outcomes in the Avon Longitudinal Study of Parents and Children: a population-based birth cohort. J Allergy Clin Immunol . 2016;138(4):1060-1070.e11.)

Translating evidence from preschool wheeze in epidemiologic studies to a clinically useful phenotypic definition led the European Respiratory Society Task Force to propose discrimination between episodic viral wheeze (EVW) and multitrigger wheeze (MTW), the latter being more likely to be associated with the development of asthma, and should be treated with inhaled corticosteroids. However, these phenotypes are not stable over time and a bronchoscopic study including children with EVW and MTW compared with nonwheezing controls showed no differences in age of onset, symptom duration, prevalence of atopy, or markers of airway inflammation (eosinophils, epithelial loss, and basement membrane thickness) between the two wheezing phenotypes, suggesting they were part of the same spectrum of disease rather than distinct pathological subtypes of asthma. In contrast, invasive studies have shown differences in bronchoalveolar lavage and endobronchial biopsy cellularity between EVW, predominantly neutrophilic, and MTW, classically eosinophilic, and there is evidence that EVW has less severe airway obstruction and impairment of gas mixing and lower FeNO compared with MTW. In practice, these features may be difficult to establish clinically, and a review of the previous Task Force guidance acknowledges greater uncertainty in the distinction between early life wheezing phenotypes and their responses to pharmacological treatment, although the presence of eosinophilia is likely to be associated with a beneficial response to inhaled corticosteroids.

Inflammatory Subtypes

The classification of asthma in adults has evolved around observed biomarkers of disease, including histopathological examination of airway mucosal inflammatory subtypes. Similar approaches in children have largely been limited to specific groups that have either been sampled opportunistically during anesthesia for unrelated conditions, or have presented with diagnostic challenges or severe disease that mandated bronchoscopic investigation. Thus, such approaches have mainly been applied to the classification of more severe disease phenotypes. Less invasive markers of airway inflammation are available, including the measurement of FeNO, the assessment of inflammatory cells in induced sputum, and the measurement of inflammatory markers in exhaled breath condensate. These biomarkers have largely been evaluated in the context of asthma control or responses to specific interventions, although application to different phenotypes of asthma in children has yielded some interesting findings. Sputum examination in a subset of the UK Isle of Wight cohort showed eosinophilia in adolescent-onset disease, which was not seen in persistent asthma from the age of 10 years, although the significance of this was not clear. As expected, sputum eosinophilia was greater in atopic compared with nonatopic asthma in this population. However, one of the potential pitfalls of using sputum cell counts to categorize asthma is that they may not be stable over time. Raised FeNO has been associated with longitudinal wheezing phenotypes, but only those associated with allergic sensitization ; therefore, it has little utility in adding to the understanding of the nature of these phenotypes. Examination of peripheral blood mononuclear cells has revealed differences in immune regulatory mechanisms between nonallergic and allergic asthma, and cytokine expression patterns in response to house dust mite extracts have displayed distinct immunophenotypes associated with house dust mite sensitization and asthma. The development of biomarkers that can be acquired through less invasive means and their implementation through systems biology is likely to lead to a better understanding of disease processes underpinning the phenotypic heterogeneity of asthma in children and may allow the development of personalized treatments or prevention strategies.

Severe Asthma

Children with severe asthma share some distinct features, including more allergic sensitization and poorer lung function than those with mild-to-moderate disease. They contribute a disproportionate burden to health care resources and are at greater risk of severe exacerbations and death from asthma. As with other forms of asthma, there is recognized phenotypic heterogeneity in this group as well. Cluster analysis of children in the US Severe Asthma Research Program (SARP) identified four different phenotypes that were characterized largely by differences in age of onset and lung function, which have been combined with adult clusters from the same study to form a composite theoretical model of severe asthma phenotypes. However, even within the adult clusters based on clinical variables, there was evidence of substantial heterogeneity in sputum inflammatory cell populations. Clusters identified in the SARP were replicated in an independent cohort, and a separate cluster analysis of children with persistent asthma enrolled in the CAMP identified five clusters that were differentiated on the basis of the atopy, airway obstruction, and exacerbation history. There was also evidence of differential responses to inhaled corticosteroids between clusters. Five clusters of severe asthma were also identified in 6- to 11-year-old children participating in the Epidemiology and Natural History of Asthma: Outcomes and Treatment Regimens (TENOR) study, in this case characterized by atopy, sex, ethnicity, and exposure to tobacco smoke. Similar approaches to disaggregating different phenotypes of severe asthma have been applied to European populations. In a French study, two severe and one milder phenotype of asthma were identified by cluster analysis in 6- to 12-year-old children. The two severe clusters included children who were highly atopic with mild lung function deficits and with poor control despite high-dose inhaled corticosteroids in one cluster, and slightly older children with severe airway obstruction in the other. Although no airway inflammatory markers were included, the former group had high peripheral eosinophil counts while the latter featured a more neutrophilic pattern. The predominant inflammatory cell type identified in the airways of children with severe asthma is the eosinophil, which has been identified in the airways of children with severe wheeze in the preschool age group. In contrast with adults, neutrophilic inflammation does not appear to predominate in severe asthma in children. There appears to be some consistency between phenotypes emerging from unsupervised cluster analyses focused on children with severe asthma and those derived from latent structures in population-based data in that the persistence of symptoms with severe exacerbations accompanied by evidence of often multiple allergic sensitization and decrements of airway function marks those children that constitute the highest burden of asthma in the pediatric population. Greater understanding of the endotypes associated with this phenotype is expected to lead to advances in the management of this often difficult to treat group.

Atopic and Nonatopic Asthma

The most common clinically applied method to phenotyping asthma is to classify it as atopic or nonatopic on the basis of evidence of allergic sensitization. Atopic asthma is characterized by eosinophilic airway inflammation, and is associated with other allergic disorders, such as eczema and hay fever, and tends to be more severe than nonatopic asthma. Estimates vary about the proportion of asthma in children that is associated with atopy; up to half of asthma cases in population-based studies are classified as nonatopic. There is a difference in the relative prevalence of atopic and nonatopic asthma between countries. The population-fraction of asthma attributable to atopy has been reported to be about twice as high in affluent compared with nonaffluent countries, and in Latin America, only a small proportion of asthma is associated with atopy. In children living in a poor area of Ecuador, the population-attributable fraction of recent asthma symptoms was only 2.4%, with heavy parasitic infestation strongly inversely associated with atopic wheezing. Therefore, asthma prevalence in countries with high microbial and parasitic exposure may be explained by different factors than those identified in affluent countries, where atopic asthma is more likely to predominate. It has been speculated that microbial load could be a risk factor for nonatopic asthma. There is certainly evidence that microbial exposure, even in affluent settings, can be differentially associated with asthma severity in children with atopic and nonatopic asthma. However, whether this mechanism is a critical pathway for asthma inception in less affluent countries remains to be conclusively determined. Associations with other putative risk factors for asthma have also been shown to have differential effects depending on the atopic phenotype. These are considered under the various categories of exposures below. It might be supposed, given the differences in exposures associated with increased risk of either atopic or nonatopic asthma, that these phenotypes would associate with clear differences in airway inflammation.

A cross-sectional study of sputum inflammatory cells and chemokines from children with atopic or nonatopic asthma based on the presence or absence of a positive skin prick test or specific serum IgE response showed that those with atopic asthma had a higher proportion of eosinophils than nonatopic asthma or nonasthma controls; however, there were no differences in sputum neutrophil percentages between the three groups, although the children with nonatopic asthma had higher total neutrophil numbers. Sputum cytokine concentrations were consistent with these findings, with atopic asthma (rather than nonatopic asthma) being associated with higher levels of interferon-gamma (IFN-γ), interleukin (IL)-2, IL-4, and IL-5, and both asthma groups having higher concentrations than the controls. Although this study seems to confirm eosinophilic airway inflammation as the typical inflammatory phenotype of atopic asthma, others have reported neutrophilic inflammation in induced sputum to predominate in nonatopic asthma. However, in a bronchoscopic study, no differences were reported in eosinophils and eosinophil cationic protein levels between children with atopic or nonatopic asthma, and there were no differences in neutrophils or related Th1 mediators IL-8 and tumor necrosis factor (TNF)-α between the asthma and the control groups. Therefore, there remains uncertainty about the inflammatory endotype underpinning asthma phenotypes in children differentiated by the presence of evidence for allergic sensitization. Additionally, concepts of allergy are changing with recognition that different patterns of sensitization are associated with different manifestations of asthma (see section “Allergy, Asthma, and the Allergic March,” p. 663 for more details).

Genetics

Changes in the prevalence of asthma over time in the latter part of the 20th century in many affluent countries have been ascribed to likely changes in environment and lifestyle; these were thought to be too rapid to be due to genetic shifts in the population, although asthma is a highly heritable disease. Twin studies were the classic design to apportion disease inception to nature (genetics) or nurture (environment and lifestyle). Monozygotic twins share the same DNA, whereas dizygotic twins will share approximately 50% of their genes on average. In the majority of cases, their early life environment will also be shared, so comparing disease outcomes between the two biological categories of twins it is possible to estimate what proportion of disease is heritable. Using this approach, a large number of twin studies have been carried out to estimate the contribution of heritable factors to asthma, and heritability estimates have generally been high, ranging from 50% to 90%. A study of over 20,000 Danish twin pairs recently not only confirmed the importance of heritable factors for asthma in childhood, but also that the importance of this diminished through the life course into late adulthood. They also reported that different genes influenced the liability to asthma in males and females, although the proportion of asthma explained by genetic factors did not differ between the sexes. A Swedish study of over 25,000 twin pairs recently reported a heritability estimate for childhood asthma of 82%. Therefore, it is clear that genetic factors are important in the development of childhood asthma. This does not negate the importance of environment in explaining temporal trends in asthma prevalence but it does highlight the roles of gene–environment interactions in asthma, and the influence of environment on gene expression through epigenetic mechanisms.

Methods used for the discovery of asthma genes have evolved over time, driven—to a large extent—by advances in genotyping technology that have scaled up throughput and reduced costs to make it feasible to apply these technologies to population studies. The earliest studies used a candidate gene approach, which is usually applied to a case-control study, to look for enrichment of a marker for a gene of interest in cases compared with controls, either without disease or from a general population sample. In this approach, the function of the gene is known a priori and, unsurprisingly, genes in immune pathways have featured large in candidate gene studies of asthma. Although this is a way to test hypotheses about association of genes with asthma, it does not have any utility for gene discovery. Many candidate gene studies for asthma have been published and extensively reviewed with particular focus on those that have been replicated in more than a handful of studies. Next came genome-wide approaches, initially through family linkage studies looking for cosegregation of genes with disease in family members, and positional cloning followed by GWAS in which the idea is to tag all common variations in the human genome and perform a hypothesis-free analysis of association of these SNP tags with disease. This has been made possible by high throughput genotyping platforms that type over 1 million SNPs. Through the HapMap project (the database has been retired by The National Center for Biotechnology Information) and the 1000 Genomes study ( http://www.1000genomes.org/ ), it is possible to impute SNPs that were not directly typed. Due to of the number of tests being carried out, the false discovery burden of GWAS means that very large sample sizes are needed to have statistical power to detect “significant” associations. One of the limitations of GWAS is that it will detect only common risk variants and most results from GWAS studies in asthma to date have yielded risk alleles that together account for a small proportion of the genetic risk, leading to the so-called “hidden heredity” of asthma. Resequencing studies are now being done to investigate rare variants in asthma susceptibility. Rare variants with large effects have been postulated to contribute to the hidden heritability of asthma, although recent evidence suggests that this may not be as important as previously thought. A meta-analysis of 20 genome-wide linkage studies identified two regions (2p21-p14 and 6p21) with evidence of linkage to asthma in European families. Large GWAS of asthma in children and adults have identified SNPs associated with asthma at genome-wide levels of significance, which have been replicated in independent populations. The association of asthma with SNPs in the 17q21 region was shown to be primarily with asthma of childhood origin. This has been confirmed in a recent GWAS of age of asthma onset with additional regions (9p24 and 17q12-q21) which was also associated with the earlier onset of asthma. Other asthma phenotypes have been considered, and GWAS have discovered associations with the asthma, hay fever/rhinitis phenotype, and childhood asthma with severe exacerbations. In the latter case, the susceptibility locus CDHR3 was subsequently found to mediate binding and replication of human rhinovirus (HRV) C.

Gene–Environment Interactions

Knowledge that environmental factors and the timing of exposure are important associations of asthma onset in early childhood has focused attention on interactions between genetic and environmental influences explaining variations in asthma risk. One consequence of assembling the very large consortia needed for GWAS studies is that they tend to come from widely varying geographical distributions and hence a very heterogeneous exposure background, which could reduce the chances of detecting genetic associations in GWAS. Most gene–environment interaction studies to date have taken a candidate gene approach, seeking to find evidence that genetic variation in known biologic pathways or with known associations with asthma modifies the effect of an exposure on asthma outcomes. These have included interactions with infectious agents, notably genetic variants in the CD14 gene and endotoxin exposure, and interaction between HRV infection and 17q21 genetic variants, a risk locus discovered in a GWAS of asthma. Following the discovery that 17q21 SNPs were associated specifically with childhood onset asthma, an interaction between variants at this locus and tobacco smoke exposure was reported. A number of other candidate gene studies of genetic interactions with tobacco smoke exposure on the risk of asthma or asthma exacerbations have been published. These have shown interactions of various genetic regions with prenatal and/or postnatal tobacco smoke exposure and asthma risk. Gene–environment interaction can also be used in a nontargeted way to identify new genetic associations with asthma. A genome-wide interaction study (GEWIS) of interactions with tobacco smoke exposure in a European consortium of birth cohorts found evidence for interaction between in utero and early childhood exposure with novel genetic variants. Interestingly, this approach did not identify interactions with variants in genes previously reported to interact with tobacco smoke exposure in the etiology of asthma, including TNF, GSTP1, and ADAM33. Another GEWIS of children from farming communities in Europe, where endotoxin exposure has been associated with protective effects on asthma and allergy, found no strong evidence for new or previously reported polymorphisms interacting with farm exposure on asthma risk. This included SNPs that had previously been reported in association with asthma or had shown interaction with farming exposure. As with GWAS, which are essentially hypothesis free, there is a high statistical penalty for multiple testing; therefore, large studies are required to have sufficient power to detect interactions, especially for polymorphisms that have a low minor allele frequency.

Other exposures for which evidence of gene–environment interactions related to asthma risk have been found in candidate gene studies include the outdoor environment with genes in antioxidant and inflammatory pathways pollutants. In the indoor environment, interactions between gas cooking and glutathione-S-transferase M1 (GSTM1) null mutations were described in relation to bronchial responsiveness in adults. A study of indoor mold exposure found no evidence of interactions with GSTP1 mutations in childhood asthma. (See Table 42.1 for details of candidate genes implicated in gene–environment interactions.)

Epigenetics

Epigenetics refers to changes in the way genes are expressed as opposed to changes in their structural sequence. This provides a mechanism by which the ability of cells to read a genetic code can be open or silenced, effectively controlling whether genes are active or dormant in a cell. A number of modifications determine gene expression, including DNA methylation, chromatin remodeling, histone modification, and the action of noncoding RNAs. Of these, the most studied to date is DNA methylation, which is technically the most straightforward to assay using current technology. Therefore, epigenetic modification is a phenotype, which can indicate responses to environmental exposures or stressors, providing a link between exposure and gene expression, and which can be transmitted to daughter cells making such modifications potentially heritable across generations (transgenerational epigenetic inheritance). Studies of DNA methylation of relevance to asthma include those which have studied epigenomic changes in association with pertinent exposures, such as tobacco smoke exposure and air pollutants, and intergenerational studies linking ancestral exposures to disease in subsequent generations. Tobacco smoke exposure during pregnancy has been associated with global DNA methylation and gene-specific effects in neonatal or cord blood, which can persist into childhood. Air pollution exposure has been associated with effects on DNA methylation in infants’ cord blood, which was dependent on the type of pollutant and timing of exposure during pregnancy. In inner-city children with asthma, DNA methylation of specific gene loci in peripheral blood differed from control children from the same environment. Mammalian cell DNA methylation is reprogrammed on a global scale at two points in the life cycle: at fertilization of the zygote and in primordial germ cells ; however, there is still uncertainty about the heritability of epigenetic marks between generations. There have been studies of the relationships of grandparental exposures with outcomes in second-generation offspring. Grandmothers’ smoking during pregnancy has been associated with asthma in her grandchild, even when the mother did not smoke during pregnancy. In a UK cohort, this association was limited to paternal grandmothers’ smoking, and was stronger for female than for male descendants. The DNA methylation in the grandchildren was not measured in either of these observational studies, but in the Norwegian Mother and Child Cohort, grandmothers’ smoking when pregnant with the mother was not associated with altered grandchild cord-blood DNA methylation at loci, which was previously reported to be associated with maternal tobacco smoking during pregnancy. The availability of high throughput arrays will shortly enable epigenome-wide association studies (EWAS) to be carried out in large case control studies of children with asthma using a similar hypothesis-free approach to GWAS.

Pregnancy and Childbirth

Early Childhood

Infections

Viral wheezing illness in early life is common, and affects around one-third of infants and young children. Viruses are also well recognized as the most frequent triggers of asthma in both adults and children. The two virus species that have received the most interest in relation to asthma development in early life are respiratory syncytial virus (RSV) and HRV. The growing interest in HRV infection is associated with modern molecular techniques of viral identification as it grows poorly in standard viral culture. RSV is a common cause of hospitalization for bronchiolitis during infancy and has long been recognized to be associated with postbronchiolitic wheezing. Serological evidence suggests that the vast majority of young children have encountered RSV infection by 2–3 years of age. So the question has been whether continued wheezing illnesses and subsequent diagnosis of asthma after bronchiolitis in infancy indicates a causal role for RSV in asthma development, or that hospitalization for RSV bronchiolitis is selective of those infants with either or both airway and immunological developmental features that already predispose them to later wheezing and asthma. A series of follow-up studies of a Swedish case control study of infants who were hospitalized with severe RSV bronchiolitis in their first year has shown an increased risk of asthma and allergic sensitization in the bronchiolitis group compared with healthy controls. This cohort has now been followed to age 18 years and the early findings have persisted. Young adults who were hospitalized for RSV in infancy compared with controls had increased prevalence of asthma and allergic sensitization, lower FEV ₁ /FVC, and increased airway responsiveness. Shorter-term cohort studies have also reported an increased risk of wheezing, asthma, and lower FEV ₁ following hospitalization for RSV bronchiolitis in infancy. In an observational study of a randomized trial of montelukast after bronchiolitis, a family history of asthma, aeroallergen sensitization, and the severity of RSV infection were predictors of asthma at age 6 years. It is possible that differences in outcomes from previous cohort studies that followed children hospitalized for bronchiolitis in infancy, but reported no increased risk of asthma, could be explained by the inclusion of a greater proportion of milder cases in population-based samples compared with clinical samples. The familial association of asthma with severe RSV bronchiolitis has recently been explored in genetic epidemiologic studies, showing a number of shared genes associated with the severity of response to RSV infection and asthma risk. This evidence provides a plausible causal link between a severe RSV lower respiratory infection in infancy and an increased asthma risk in later childhood. Empirical evidence of differences in immune responses to RSV infections and subsequent asthma has also been provided by studies of blood and airway chemokines and cytokines during infection in relation to subsequent asthma. Gene variants in ILRL1, an asthma-associated gene, and ILRL1-a expression in the upper airway have been reported in association with severe RSV bronchiolitis. Increased expression of CCL5 (RANTES) in the upper airway at the time of severe RSV bronchiolitis, in addition to the development of aeroallergen sensitization, has been linked to an increased risk of subsequent asthma. Thus, there is a shifting focus to innate and adaptive host responses and their dysregulation in the etiology of asthma, and therefore the therapeutic potential of immunomodulatory interventions, such as vaccines in primary prevention. Meanwhile, trials of therapeutic interventions with antiinflammatory medications after bronchiolitis have been generally disappointing. Treatment with long-term, high-dose inhaled corticosteroids following hospitalization for RSV bronchiolitis has not been shown to influence respiratory outcomes. Systematic reviews of leukotriene inhibitors after bronchiolitis in young children show that they may reduce symptoms of postbronchiolitic wheezing, but there is no evidence of a beneficial effect on the longer-term outcomes of recurrent wheezing or the use of inhaled corticosteroids for asthma. The macrolide antibiotic azithromycin has been shown to reduce neutrophilic airway inflammation in a murine model of viral bronchiolitis and has been shown to reduce the risk of subsequent severe lower respiratory infections in children with a history of such infections. In a proof-of-concept trial, 14 days treatment with azithromycin after bronchiolitis was shown to reduce upper airway levels, but not serum IL-8 levels, and to reduce respiratory symptoms in the succeeding 12 months after illness. Immunoprophylaxis for RSV with palivizumab is currently restricted to infants at high-risk of severe RSV bronchiolitis, so results may not be generalizable to the general population. Short-term reductions in the frequency of wheezing symptoms have been reported in palivizumab-treated preterm infants and, although suggestions of lower asthma prevalence in association with better adherence to palivizumab were reported in a retrospective analysis of an eligible cohort, this evidence needs to be confirmed in further studies and in term infants at high risk for asthma.

A complex interplay between early life viral infection, host immune responses, and interactions with allergic sensitization in the genesis of asthma has also emerged from studies of the relationships between HRV infection and subsequent asthma risk. HRV is the most common virus identified in early life wheezing illnesses. The association of early HRV-associated severe respiratory illnesses and subsequent wheezing and asthma was initially reported in a Finnish observational cohort study followed to 11 years of age. Prospective birth cohort studies have since confirmed this association. The Childhood Onset of Asthma (COAST) study in the US has been instrumental in uncovering the mechanisms underlying this strong association ( http://www.medicine.wisc.edu/coast/family ). A recent follow-up of this high-risk birth cohort has shown a strong positive association between HRV and not RSV with regard to wheezing in the first 3 years and asthma at age 13 years ( Fig. 42.5 ). There was also evidence that the presence and timing of aeroallergen sensitization was associated with subsequent asthma in this cohort; 65% of those sensitized in the first year compared with 40% sensitized by 5 years but not in the first year and 17% not sensitized by age 5 years, had asthma at 13 years and there was an additive effect with HRV wheezing illnesses. Furthermore, a temporal analysis of the association between allergic sensitization and viral wheezing illnesses in the COAST study has shown that allergic sensitization increases the risk of wheezing illnesses caused by HRV; however, the converse, HRV wheezing preceding allergic sensitization, was not supported. This suggests a causal role for allergic sensitization in the biological pathway of asthma development following respiratory infection, particularly with HRV. Blunting of the seasonal increase of asthma exacerbations associated with viral infections by pretreatment with omalizumab provides additional support for the pivotal role of allergic sensitization in modifying the response to viral infections in the asthmatic airway. The risk of subsequent asthma following HRV infection has also been found to be enhanced by genetic variants at the 17q21 locus, an asthma associated locus in the GWAS of asthma in children, and with expression of two of the genes at this locus, ORMDL3 and GSDMB in the COAST and the Copenhagen Study on Asthma in Childhood (COPSAC) birth cohorts. It is likely that dysregulated immune responses to viral infections of the airway play an important role in asthma inception. Candidate genes selected for their role in antiviral responses have been identified in association with viral-induced asthma exacerbations and childhood asthma phenotypes. Impaired innate immunity with deficient interferon responses to respiratory viral infections has been previously reported. Studies of airway epithelial cells from children with asthma compared with cells from healthy controls have shown increased susceptibility of the former to HRV infection with increased viral replication, reduced production of interferons, and impaired apoptotic responses, although it is conceivable that the presence of asthma alters cellular immune responses rather than arising as a direct consequence of immune dysregulation.

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Drug Administration by Inhalation in Children
2. Pneumonia in Children
3. Lung Injury Caused by Pharmacologic Agents
4. Pediatric Lung Transplantation

Stay updated, free articles. Join our Telegram channel

Join