OUTCOME OF PRESCHOOL ASTHMA

RELATED PAGES:
PreSchool Asthma
Pediatric Asthma
Role of Allergy in Asthma

Page prepared April 2005

Cough and wheeze in the preschool years are said to occur with respiratory tract infections in 30% of preschool children. It has been suggested that children who are non-atopic (i.e. not allergic) will commonly outgrow this problem with age while those who are allergic, will tend to have persisting asthma. Allergy is the primary mechanism for creating eosinophilic inflammation in the airways of children and eosinophilic inflammation is considered to be the main pathophysiologic mechanism in asthma which leads to airway remodelling and chronic asthma.

A number of groups from around the world have enrolled wheezing preschool or school children in long-term studies to follow them and document the outcome while defining risk factors in infancy that will enable identification of the children likely to have persisting asthma.

In these pages, I will review the literature with the follow-up publications from these groups.

COHORT OUTCOME STUDIES CHILDHOOD ASTHMA

Initially I will discuss two cohort studies that enrolled children in early school-age and have followed the subjects to adulthood, the Melbourne cohort and the Dunedin cohort. These studies have contributed a lot to our understanding of the role of allergy in childhood asthma.

SCHOOL-AGED CHILDREN:

MELBOURNE COHORT:
One of the earliest cohort studies of children with asthma was initiated in Melbourne in 1964 when a group of children were randomly selected from a birth cohort of 7 year old children and entered for study. That study did not actually examine preschool children and at entry, the authors concluded that there was no difference in children who had been diagnosed as wheezy bronchitis or asthma. However, the wheezy bronchitis group was less atopic and had lower eosinophil counts than the asthmatic groups. Those authors had suggested that there was no difference between "wheezy bronchitis" and asthma but the differences in atopy and peripheral eosinophils would now be interpreted as a significant difference between the groups. Moreover, it was found that the original random selection of patients from the birth cohort, failed to enter enough children with severe asthma for follow-up so that at age 10, a selected group with severe wheezing were added for follow-up. 
Williams H, McNicol KN 1969

These subjects were followed prospectively every 7 years and had their last examination at age 42, in 1999 when 87% or the original cohort was restudied. Phelan 2002. It was found that the majority of children with wheezing with viral illnesses (many of the wheezy bronchitis group) had a benign course and many were no longer wheezing as adults. In contrast, the children with asthma especially in the added population of severe wheezing, continued to wheeze as adults. Changes in lung function had occurred by 14 years of age. Those subjects with mild asthma had no significant loss of lung function.

I interpret these studies to indicate that there is such a thing as "wheezy bronchitis" and it is now usually referred to as viral-induced wheezing in preschool children. Many of those children would have outgrown the wheezing before age 7 and not even be enrolled in the study. In contrast the group of severe asthmatics entered at age 10 would likely be atopic asthmatics but it is not clear whether they began wheezing in the preschool age as persisting asthma (as defined by Martinez et al.) but even in the Melbourne study, they represent the persisting asthma and could be recognized on cross-section at admission to the study at age 10. That suggests that in a doctor's office, if a child at age 10 is atopic and has difficult asthma, they are likely to have asthma that will persist into adulthood and they may already have changes in lung function.

DUNEDIN COHORT:
One of the cohorts of children which has proven extremely useful to the understanding of childhood asthma was initially used to determine the prevalence and severity of childhood asthma in New Zealand and consisted of 815 children from a birth cohort studied by questionnaire, clinical examination and pulmonary physiological measurements at age 9 years.  Jones 1987 When these children were followed to 13 years of age it was found that "sensitivity to house dust mite and to cat dander were highly significant independent risk factors associated with the development of asthma (whether defined as recurrent typical respiratory symptoms, increased airway responsiveness, or the concurrent presence of both), whereas grass sensitivity was not a significant independent risk factor for asthma." This cohort of children was not actually studied in the preschool age and any history obtained for that period would have relied on the memory of the parents which is subject to bias.  Sears 1989 These children have now been followed into adulthood (i.e. from age 9 to 26) and the risk factors identified. Sears 2003 The authors concluded that "In an unselected birth cohort, more than one in four children had wheezing that persisted from childhood to adulthood or that relapsed after remission. The factors predicting persistence or relapse were sensitization to house dust mites, airway hyperresponsiveness, female sex, smoking, and early age at onset. These findings, together with persistently low lung function, suggest that outcomes in adult asthma may be determined primarily in early childhood."

This study confirmed the importance of allergy, in this instance sensitivity to house dust mites, to the persistance of asthma and the relapse of asthma when the children seemed to have "outgrown" it. The data also suggested that lung function abnormalities seen in adult asthmatics may have begun in childhood. Since the study entered the children at age 9, their preschool history is unknown so that the cohort cannot be easily compared to the Tucson cohort (described below) that followed the children from infancy.

NEW SOUTH WALES:
This cohort (a random sample of 718 children) was enrolled at ages 8 - 10 years in 1982 and from 1984 - 1992, five follow-up surveys were performed every 2 years. A seventh survey was performed between 1997 and 1999 encompassing 498 subjects. The purpose of the study was to to examine the onset and remission of atopy and asthma during adolescence and to evaluate potential risk factors.  Xuan 2002  Atopy, airway hyperresponsiveness (AHR), and wheeze in the last 12 months were measured at each survey. Late onset, remission, and persistence were defined based on characteristics at the initial survey and the changes in characteristics at the follow up surveys. The authors concluded that the onset of AHR was uncommon during adolescence, but the risk of acquiring atopy and recent wheeze for the first time continued during this period. Atopy, particularly when present at the age of 8–10 years, predicted the subsequent onset of wheeze.

Conclusions: These studies proved that atopy was a major risk for the development of asthma in children and the Dunedin study suggested that sensitivity to indoor, year-round allergens such as mites, danders and moulds were more important than pollens for creating asthma in children. However because the studies enrolled children of school age, it was not clear if the results applied to infants and preschool aged children.

The next three cohort studies to be described enrolled children in infancy in order to determine the earliest risk factors for wheezing in children.

INFANT STUDIES:

TUCSON COHORT
Martinez et al. initially described the heterogeneity of preschool asthma and separated the children into several groups followed from infancy to age 6 Martinez 1995. They described 4 groups:

  1. transient wheeze - children born with smaller than normal airways often associated with smoking mothers.
  2. early onset wheeze
  3. persisting wheeze
  4. late onset wheeze

It was found that the transient wheezers had decreased airflow compared to normal infants but "outgrew" the problem within a few years. Those infants and the early onset wheeze were not associated with a family history of allergy and asthma and had normal serum IgE levels. The early onset asthma had normal airflow in infancy and was also not associated with a family history of allergy and asthma. They tended to stop wheezing by the follow-up at age 6. In contrast, the persisting wheeze and late onset wheeze were associated with a family history of allergy and asthma and tended to have elevated IgE levels in infancy. When these children were first studied, they did not have tests done for specific allergens either by skin testing or RAST testing. At follow-up at age 6 the children had skin testing performed.

There have been numerous publications examining different aspects of the Tucson cohort attempting to determine the important relationships between preschool wheezing and persisting asthma. One of the most important publications was the development of an algorithm to predict which wheezing preschool child might fall into the persisting asthma category.  Castro-Rodriguez 2000  A stringent index included frequent wheezing during the first 3 yr of life and either one major risk factor (parental history of asthma or eczema) or two of three minor risk factors (eosinophilia, wheezing without colds, and allergic rhinitis). The index was developed after the cohort of children had been followed and studied to age 13. A loose index was also developed and it required any wheezing during the first 3 yr of life plus the same combination of risk factors described previously. The authors found that 59% of children with a positive loose index and 76% of those with a positive stringent index had active asthma in at least one survey during the school years. Over 95% of children with a negative stringent index never had active asthma between ages 6 and 13.

The data suggested that the likelihood that wheezing preschool children less than age 3 would have persisting asthma could be predicted using the index. However there was a problem with the index in that it failed to give sufficient weight to atopy because in the initial 1995 publication, the preschool children entered into the study were not tested for specific allergy. Allergic rhinitis was chosen as a proxy risk factor representing atopy but allergic rhinitis is difficult to diagnose in preschool children since many of them have constant nasal discharge from frequent colds. As will be described, in cross-sectional studies that I performed, we identified that allergic children with wheezing and sensitivity to aeroallergens at a young age were more likely to require maintenance inhaled steroid for their asthma management than the non-atopic preschool wheezers. More recently the PEAK trial, which is currently underway, modified the Tucson index to include positive skin test to aeroallergens as a major criteria and positive skin test to food as a minor criteria along with the family history. This modification of the index recognizes that wheezing preschool children who are not atopic are less likely to have persisting asthma than preschool wheezing children who are allergic.

MANCHESTER COHORT
In this study, "The Manchester Asthma and Allergy Study", the goal was to determine the relationship of lung function and atopy to children who were stratified into the same wheezing phenotypes as described by the Tucson group Lowe 2005 . The study was a population-based birth cohort in which participants were recruited prenatally and attended review clinics at age 3 and 5 years. A standard respiratory questionnaire was interviewer-administered at age 3 and 5 years to collect the information on symptoms, physician-diagnosed illnesses, and treatments received. According to parentally reported history of wheeze at two follow-up evaluations, children were assigned to the following categories:

sRaw was measured at age 3 and 5 years using a whole-body plethysmograph and served as a measure of lung function in these preschool children. Allergic sensitization (mite, cat, dog, grasses, milk, egg) was ascertained at age 3 and 5 by skin-prick testing. In the follow-up study, published in 2005, 874 children, 530 (60.6%) successfully performed sRaw measurement. A total of 496 (93.6%) of these 530 children attended the 5-year follow-up (248 never wheezed, 115 had transient early wheeze, 22 developed late-onset wheezing, and 78 had persistent wheezing). The authors found that the sRaw measurement did not differ between children who had never wheezed, children with transient wheezing and children with late-onset wheeze. The authors concluded that increasing sRaw value (OR 5.5, 95% CI 1.2–25.9; p = 0.03) and child's sensitization (OR 2.8, 95% CI 1.3–5.8; p = 0.008) were significant and independent associates for increasing the the risk of persistent wheezing.
This study helped to confirm the usefulness of the categories of preschool wheezing described by the Tucson group and by implication, the algorithm for predicting persisting wheeze in preschool children.

ISLE OF WIGHT
Children were seen at birth, and at 1, 2, 4, and 10 years of age in a whole-population birth cohort study (1,456 subjects). Information was collected prospectively on genetic and environmental risk factors. Skin-prick testing was performed at 4 years of age. The purpose of the study was to identify early life factors (ie, first 4 years) associated with wheeze, asthma, and bronchial hyperresponsiveness (BHR) at age 10 years, comparing their relative influence for these conditions.  Arshad 2005. They found Asthmatic heredity, predisposition to early life atopy, plus early passive smoke exposure and recurrent chest infections are important influences for the occurrence of wheeze and asthma at 10 years of age. They also found that for bronchial hyperreactivity, atopy was the strongest factor.

GERMAN MULTICENTRE ALLERGY STUDY (MAS):
The German MAS study Illi 2001, Lau 2002 was designed "to prospectively investigate the pattern of atopic sensitization typically associated with the development of asthma in childhood". The study followed 1314 children from birth to the age of 7 years using parental questionnaires on asthma and asthmatic symptoms which were completed 6 times up to the age of 2 years and from then on yearly. Specific IgE to 9 food and inhalant allergens was determined yearly. At the age of 7 years, pulmonary function testing and a bronchial histamine challenge was performed in 645 children. It was found that children who had asthma at age 7 had become sensitized to allergens significantly earlier than children who were allergic but did not have asthma (39.4% before age 1 year vs 21.0%, P =.015). Early atopic sensitization without any sensitization to inhalant allergens at the age of 7 years conferred no increased risk for asthma at this age. The authors concluded that atopy itself did not create wheezing but they were independent phenomenon. Only those children sensitized to any allergen early in life and sensitized to inhalant allergens by the age of 7 years were at a significantly increased risk of being asthmatic at this age (odds ratio, 10.12; 95% CI, 3.81-26.88). The study also examined the rate of atopic dermatitis in this birth cohort and its association with asthma. Illi 2004. Atopic dermatitis was significantly associated with early onset of wheezing and a specific sensitization pattern that included some foods but cat and mite at less than 2 years of age. The authors concluded that there was no progression from atopic dermatitis to wheeze but they developed concomitently and such children had significantly reduced lung function at age 7 compared to those children who did not have atopic dermatitis and early wheezing.

RSV AND BRONCHIOLITIS

POST-BRONCHIOLITIS AND RSV:
It has been known for a long time that following bronchiolitis about 40% of the children develop recurring wheezing with colds. It has not been clear whether the RSV creates the respiratory problem or simply selects a group of children who are susceptible to lower respiratory tract infection. I will not review the entire literature on RSV bronchiolitis but will select some publications which seem to bear on the issue of the outcome of preschool wheezing. It seems likely that the outcome is different for RSV-induced bronchiolitis when the patient population studied are drawn from a community-based population versus a population who had been hospitalized for severe RSV-induced bronchiolitis.

Tucson cohort: The patient population followed in the Tucson cohort were examined for the outcome of RSV-induced bronchiolitis Stein 1999  The authors concluded "RSV lower respiratory tract illnesses in early childhood are an independent risk factor for the subsequent development of wheezing up to age 11 years but not at age 13. This association is not caused by an increased risk of allergic sensitisation." In this birth cohort of mild, non-hospitalized RSV-induced bronchiolitis, the children who continue to wheeze with colds, are generally not atopic and outgrow the problem by age 13. This is a different outcome compared to the studies enrolling infants hospitalized with bronchiolitis where a high percentage develop atopy and persisting asthma.(see below)


SWEDISH RSV STUDY Sigurs has followed 47 infants "hospitalized" with RSV-induced bronchiolitis at a mean age of 3.5 months and compared them with 93 matched controls  Sigurs 1995 and published follow-up data at age 3 , age 7.5 years,  Sigurs 2000 and age 13  Sigurs 2005  At the initial follow-up visit at age 1, the authors concluded that Respiratory syncytial virus bronchiolitis during the first year of life apparently is an important risk factor for the development of asthma and sensitization to common allergens during the subsequent 2 years, particularly in children with heredity for atopy/asthma. At the second follow-up, at age 3, asthma, defined as three episodes of bronchial obstruction verified by a physician, was found in 11 of 47 children (23%) in the RSV group and in 1 of 93 children (1%) in the control group (P < .001). A positive test for IgE antibodies was noted in 14 of 44 (32%) of the RSV positive children and in 8 of 92 (9%) children in the control group (P = .002). The children were evaluated at 13 years of age and it was found that the occurrence of symptoms over the previous 12 months was significantly higher in the RSV group than among the control subjects, 43% versus 8% for asthma/recurrent wheezing and 39% versus 15% for allergic rhinoconjunctivitis. Sensitization to common inhaled allergens was more frequent in the RSV group than in the control subjects, judged by skin prick tests (50% versus 28%; p = 0.022), or by serum IgE antibodies (45% versus 26%; p = 0.038). Compared with the control subjects, the RSV group showed mild airway obstruction both at rest and after bronchodilation, and had slightly more reactive airways. RSV bronchiolitis in infancy severe enough to cause hospitalization is a risk factor for allergic asthma in early adolescence.  Sigurs 2005  Unfortunately, the characterization of atopy was not done at age 3.5 months and not done fully at age 1, the first follow-up. So it is not possible to compare this population with the group studied by the Finnish investigators (see below). It is not entirely clear whether the RSV selected from atopic patients destined to be asthmatic but the index group clearly has a family history of atopy and asthma which is a significant risk factor for persisting asthma identified in the Tucson algorith for predicting persisting asthma. The studies published by Sigurs confirm that infants with bronchiolitis, in this case RSV-induced, severe enough to warrant hospitalization are at increased risk for atopy and persisting asthma. In that respect these studies are similar to the results of the Finnish studies and suggest that bronchiolitis severe enough to warrant hospitalization is a marker for persisting asthma unlike community based bronchiolitis as described by Stein et al.


Swedish - Wennergren: In 1992 Wennergren et al published a study of 101 infants hospitalized with wheezing before age 2 at a median age of 10 months. By 3 - 4.5 years later when the group of subjects was re-examined, the factors which correlated significantly with persistent asthma included the need for daily medication for at least 6 months and other past or present atopic symptoms.

These subjects (92 of the original 101 children) were reinvestigated at age 10 and it was found that 30% of them continued to have symptoms of asthma at age 10 while 37% of them had a positive histamine challenge test indicating bronchial hyperreactivity.  Wennergren 1997 Persistent asthma correlated significantly to the presence of some other atopic disease and to intense obstructive disease as a young child, while initial RSV infection did not. There was no control group for comparison but the authors concluded that the prevalence of asthma was high in this population compared to the general population of Sweden.

FINNISH BRONCHIOLITIS.
One hundred consecutive infants younger than 24 months treated in the hospital for acute bronchiolitis were enrolled in a study to evaluate whether early anti-inflammatory therapy with nebulized cromolyn sodium or budesonide reduces wheezing after bronchiolitis. A significant reduction in hospital admissions for bronchial obstructions was seen in the budesonide group and in the children with atopy in both treatment groups (P < .05). The children with atopy had significantly more subsequent wheezing episodes and hospital admissions than those without atopy (P < .05). Reijonen 1996 The same patients were evaluated by measurement of serum ECP. At entry, 14 of 92 (15%) children had high (> or = 16 micrograms/L) levels of ECP in their serum. During the 16-week follow-up period, this group of patients had significantly more physician-diagnosed episodes of wheezing (86% vs. 43%, P < 0.01) and hospital admissions for wheezing (64% vs. 19%, P = 0.001) than those with serum levels of ECP < 16 micrograms/L. The authors concluded that a high serum ECP concentration during the acute phase of bronchiolitis is a specific but insensitive predictor of wheezing after bronchiolitis. It seems likely that the high serum ECP was identifying infants with atopic wheezing who were mobilizing eosinophils during the initial episode of wheezing requiring hospitalization. It is likely that the peripheral eosinophil count would also be elevated in that group identifying them as atopic infants at risk for persisting asthma.  Reijonen 1997 The authors followed this group of children and at school age evaluated the association of specific IgE antibodies to food and inhalant allergens and persisting asthma. They concluded that "when present in wheezing infants, specific IgE of >or=0.35 kU/L to wheat, egg white, or inhalant allergens are predictive of later childhood asthma. Consequently, detection of those specific IgE antibodies in wheezing infants may facilitate the early diagnosis of asthma, especially in cases with no clinically evident atopic manifestations.  Kotaniemi 2003 This patient group has been followed and studied several times.  (3 year follow-up Reijonen 2000,  , 6 year follow-up Kotaniemi-Syrjanen 2002)

The patient population was selected from infants hospitalized with bronchiolitis and this seems to have created a selection bias for atopic asthmatics. Most of the studies on RSV or bronchiolitis did not test the infants for allergy as was done in this study. The "baseline data" obtained on enrollment as an infant into the study Over 10% of the total followed (13/81) had the presence of IgE to inhalant allergens at the index admission. At less than age two, it is not common to have atopic sensitization to inhalant allergens compared to foods (Foucard 1973). This suggests that a group of atopic infants and toddlers with early conversion to inhalant allergen sensitivity and wheezing with viral infections was enrolled. Overall 36 of 81 patients had IgE to inhalant or food allergens (44%) which is high for an unselected population such as would result from an epidemiologic study of a birth cohort. The population of young infants admitted to hospital with a wheezing illness included an excess number of atopic wheezers. Without a control population it may not be possible to categorically state that "infants hospitalized with bronchiolitis" are a population selected for atopic asthma that according to the PEAK study algorithm will develop persisting asthma. Right from infancy these wheezers responded to anti-inflammatory medication suggesting that this population when seen in a doctor's office, having had an admission to hospital for bronchilitis should be regarded as high risk for persisting atopic asthma and should be treated with inhaled steroid, perhaps maintenance inhaled steroid as recommended in the Canadian Consensus Guidelines.

INTERPRETATION OF RSV AND BRONCHIOLITIS DATA: These publications suggest an interpretation for conflicting data on the outcome of RSV bronchiolitis and bronchiolitis in infants. It seems likely that when bronchiolitis is studied in community populations which are milder episodes, a proportion of the infants develop viral induced preschool wheezing with recurrence of wheeze with every cold (Stein et al.). Such children are likely to outgrow the wheezing as they get older. However when the population being studied is drawn from the more severe hospitalized group, the patients are more likely to be drawn from atopic asthmatics who will have recurring atopic asthma involving eosinophilic inflammation (see Swedish and Finnish studies above). Thus when a physician is confronted with recurring viral-induced wheeze in a patient who was hospitalized for bronchiolitis, an attempt should be made to determine if the patient is atopic and management should be the same as school-aged atopic asthmatics with the use of inhaled steroid as first-line therapy and a likelihood for needing maintenance inhaled steroid.

CROSS-SECTION STUDIES PERFORMED BY THE AUTHOR:
In general epidemiologic population studies that follow the children over time are considered much stronger evidence than cross-sectional studies that recruit patients from a clinic. That is because cross-sectional studies are subject to selection bias, that is bias in the original population who were selectively referred to the clinic and probably represent a more symptomatic group, hence the referral to the specialty clinic. It is not possible to know whether patients who were not referred to the clinic are similar to the patients who were referred therefore the population studied may not resemble the entire population of wheezing preschool children. However despite that caveat, a cross-sectional population can indicate whether a doctor in his office is likely to be able to recognize the individual patient who has the characteristics seen in the population studies.

In 1988, we published a study of school-aged childhood asthma that determined on cross-section that children that the severity of asthma correlated with the degree of allergy i.e. the more allergic children tended to have more difficult asthma requiring more medication for treatment than those who were less allergic or non-allergic. Zimmerman 1988 Allergy was assessed by numbers of skin test positive, the size of the positive skin tests and the history of allergy symptoms. We hypothesised that children with more difficult asthma were more atopic and more easily sensitized to allergens and more likely to have allergen-induced inflammation in the airways. Similarly we found that preschool children between age one and two who were more atopic, with IgE antibody to food allergens and inhalant allergens had more difficult, chronic asthma requiring chronic treatment with inhaled steroid compared to non-allergic preschool children with wheeze. Zimmerman 1988 In this population, despite the referral to a hospital-based clinic half of the children were completely non-allergic and they had milder asthma with less requirement for maintenance asthma medication. In fact in a non-hospital-based practice, the majority of preschool children with wheeze are not allergic, and have intermittent cough and wheeze with colds.

In 1993, we published that serum eosinophil cationic protein (ECP)(and peripheral blood eosinophil counts) distinguished between symptomatic and asymptomatic asthma in school-aged children with the former patients having higher levels of serum ECP than the latter. Zimmerman 1993 In a subsequent study, we examined peripheral eosinophils and serum ECP in preschool children with wheezing and found that the allergic children (those with positive skin tests) had higher peripheral eosinophil counts and serum ECP than the non-allergic children (those without any positive skin tests). Zimmerman 1994 We hypothesised "that the mildest form of childhood asthma in preschool children occurs primarily with viral infections in non-atopic children. These children would show minimal involvement of eosinophils either mobilized in the circulation or activated as reflected in the serum levels of ECP and EPX." We further hypothesised that "the patients with the mildest form of wheezing with minimal eosinophil involvement improve with age, so that most childhood asthmatics older than the age of 6 with symptomatic asthma are atopic with apparently greater eosinophil involvement". These hypothesese have been amply borne out by the much stronger population studies that have followed the course of preschool wheezing over time as previously cited. But the studies suggest that the different prognosis for preschool wheezing between atopic and non-atopic children can be recognized early for example by an algorithm published by Castro-Rodriguez et al. Castro-Rodriguez 2000  Moreover preschool wheezing children should be tested for allergy to distinguish the atopic population likely to have persisting asthma from the non-atopic children likely to outgrow the problem.

In fact in the PEAK study now underway, as will be discussed later, the Tucson algorithm has been modified to include skin testing in preschool children in order to identify the children more likely to have persisting asthma and the study is attempting to determine whether early treatment of those high risk preschool children with maintenance inhaled steroid, at a point in time when their asthma symptoms seem intermittent, could prevent changes in the airways that are associated with persistence of the asthma.

FORMULATION OF THE PROBLEM WITH MANAGEMENT OF PRESCHOOL WHEEZING.


From the NAEPP Report (Expert panel report NIH): Guidelines for the diagnosis and management of asthma, update 2002.Go NAEPP 2002. This report summarizes the nature of the problem very well.

An expert panel convened by the NIH hsd evaluated the literature on preschool children.
Section 3: PREVENTION:
"Recent research suggests that the course of asthma may vary markedly between young children, older children and adolescents, and adults, and this variation is probably more dependent upon age than symptoms."