Asthma: epidemiology, etiology and risk factors

Prenatal tobacco smoke

Prenatal maternal smoking has been consistently associated with early childhood wheezing,^22–25 and there is a dose–response relation between exposure and decreased airway calibre in early life.^26,27 Prenatal maternal smoking is also associated with increased risks of food allergy,²⁴ cytokine responses in the cord blood^28,29 and concentrations of nitric oxide in exhaled air in newborns.³⁰ Studies have shown a clear prenatal effect of smoking; this effect is increased when combined with postnatal smoke exposure.

Diet and nutrition

Observational studies examining prenatal nutrient levels or dietary interventions and the subsequent development of atopic disease have focused on foods with anti-inflammatory properties (e.g., omega-3 fatty acids) and antioxidants such as vitamin E and zinc. Several studies have demonstrated that higher intake of fish or fish oil during pregnancy is associated with lower risk of atopic disease (specifically eczema and atopic wheeze) up to age 6 years.^31–33 Similarly, higher prenatal vitamin E and zinc levels have been associated with lower risk of development of wheeze up to age 5 years.^34–36 However, no protective effect against the development of atopic disease in infants has been shown for maternal diets that excluded certain foods (e.g., cow’s milk, eggs) during pregnancy.^37–40 The authors of 2 recent studies^41,42 reported an inverse relation of maternal vitamin D levels with wheeze in early life, but no relation with atopy or symptoms in later life.

Stress

A number of animal models have suggested that prenatal maternal stress acts through regulation of the offspring’s hypothalamic–pituitary–adrenal axis to decrease cortisol levels, which may affect the development of an allergic phenotype. Although there is a correlation between caregiver stress early in the infant’s life and higher levels of immunoglobulin E in the infant^43–45 and early wheezing,⁴⁶ no studies to date have shown an association with asthma.^47,48

Antibiotic use

The association between prenatal antibiotic treatment and subsequent development of atopic disease has been examined in 2 ways: with treatment as a dichotomous predictor (i.e., any antibiotic use) and by number of courses of antibiotics during pregnancy. Longitudinal cohort studies examining any antibiotic use showed a greater risk of persistent wheeze and asthma in early childhood^49,50 and a dose–response relation between number of antibiotic courses and risk of wheeze or asthma.^49,51

Mode of delivery

Development of atopy was 2 to 3 times more likely among infants delivered by emergency cesarean section,^29,52–56 although no such association occurred with elective cesarean section.^{29,52,53,56–59} Potential reasons for these findings include maternal stress and differences in the infant’s gut microflora associated with different modes of delivery.

Phenotypes of asthma

Although some 50% of preschool children have wheezing, only 10%–15% have a diagnosis of “true” asthma by the time they reach school age.^13,60 Commonly described phenotypes in early infancy and childhood are transient wheezing, nonatopic wheezing, late-onset wheezing and persistent wheezing.⁶¹ Only transient wheezing in early infancy has been well characterized, with decreased airflow rates on pulmonary function testing at birth,^56,60,62 onset of wheezing within the first year and resolution by mid-childhood with no lasting effects on pulmonary function.

The other 3 phenotypes have been described primarily by age of onset in cohort studies, and their genesis in early infancy is largely unknown. The majority of children with persistent wheezing (in whom asthma will subsequently be diagnosed) experience their first symptoms before age 3. By 3 years, they have abnormal lung function that persists to adulthood,^13,60,61 and by adolescence, most have atopy. Of children with nonatopic and late-onset wheezing, some experience remission, whereas others experience persistent symptoms and atopy.⁶³

Distinguishing among these different phenotypes in early childhood is critical to understanding the role of risk factors and their timing in early infancy.

Breastfeeding

The influence of breastfeeding on the risk of childhood atopy and asthma remains controversial. The following represents observational data accumulated to date. Some studies have shown protection,^64–66 whereas others have reported higher rates of allergy and asthma among breastfed children.^67,68 A meta-analysis⁶⁹ and several individual studies^66,70 showed that exclusive breastfeeding for at least 3 months was associated with lower rates of asthma between 2 and 5 years of age, with the greatest effect occurring among those with a parental history of atopy. One of the difficulties in interpreting these data lies in differentiating viral-associated wheeze in childhood from development of atopic asthma. In a longitudinal birth cohort study, breastfeeding was associated with a higher risk of atopic asthma in later childhood, with the greatest in fluence occurring among those with a maternal history of atopy.^67,68,71

The influence of avoiding nutritional allergens during breastfeeding is also controversial. In some studies, exclusion of milk, eggs and fish from the maternal diet was associated with decreased atopic dermatitis in infancy,^72,73 but other studies found no association.^40,74,75 Studies following children to 4 years of age have demonstrated no effect of maternal dietary restriction during lactation on the subsequent development of atopic diseases, including asthma.⁷⁶

Lung function

Decreased airway calibre in infancy has been reported as a risk factor for transient wheezing,⁶⁰ perhaps related to prenatal and postnatal exposure to environmental tobacco smoke.^26,27 Furthermore, the presence of airways with decreased calibre has been associated with increased bronchial responsiveness and increased symptoms of wheeze.²⁶ Several studies have suggested an association between reduced airway function in the first few weeks of life and asthma in later life.^62,77 The magnitude of the effect of this risk factor in isolation (i.e., without concomitant allergy) is unclear; perhaps individuals with smaller airways require less stimulus (i.e., airway inflammation) before symptoms become apparent.

Children with wheezing (and diagnosed asthma) persisting to adulthood have a fixed decrement in lung function as early as age 7 or 9 years.^13,78 Recent studies of preschool children have documented abnormal lung function in children with persistent wheezing as young as age 3 years.⁶¹ However, some infants in whom persistent wheezing develops have normal lung function shortly after birth, which suggests a critical period of exposures within the first few years of life, before the development of these persistent abnormalities in expiratory flows.^60,79 In contrast, infants who exhibit early transient wheezing have decreased airflow shortly after birth.^60,80 Maternal smoking with in utero nicotine exposure has been correlated with this type of lung dysfunction,^26,27,60 but the effects of other exposures have been less well studied.

Family structure

Family size and the number and order of siblings may affect the risk of development of asthma. The hygiene hypothesis posits that exposure of an infant to a substantial number of infections and many types of bacteria stimulates the developing immune system toward nonasthmatic phenotypes.^81,82 This may be exemplified in the real world by large family size, whereby later-born children in large families would be expected to be at lower risk of asthma than first-born children, because of exposure to their older siblings’ infections.

Although this theory has been supported by some studies of allergy prevalence,^83,84 it has been partially refuted by recent studies of asthma prevalence suggesting that although large family size (more than 4 children) is associated with a decreased risk of asthma, birth order is not involved.^85,86 Furthermore, doubt has been cast on simplistic renditions of this hypothesis, in that infections per se cannot explain some epidemiologic patterns (e.g., prevalence rates for allergy and asthma are high in some South American countries, where exposures to infection are higher than in some countries with lower rates of asthma³). In addition, not only allergic but also autoimmune and other chronic inflammatory diseases are increasing,⁸⁷ a trend that is difficult to explain by the hygiene hypothesis alone, since allergic and autoimmune diseases are associated with competing immunologic phenotypes.

Socio-economic status

Children of parents with lower socio-economic status have greater morbidity from asthma,^88–92 but findings with respect to the prevalence of asthma are mixed.^93–97 Such results may depend both on how socio-economic status is measured and on the specific outcome examined. Some studies have reported associations of lower socio-economic status with greater airway obstruction and symptoms but not with a diagnosis of asthma.^91,92 Whether socio-economic status is as relevant to the incidence of allergy and asthma as it is to the expression, severity and management of these diseases re mains unclear. Parental stress has also been prospectively associated with wheezing in infancy,⁴⁶ and family difficulties have been linked to asthma.^48,98 Children whose caregivers report high levels of stress and who have difficulties parenting are at greatest risk for asthma.⁹⁹

Antibiotics and infections

The use of antibiotics has been associated with early wheezing and asthma in several studies,^47,100,101 One suggested mechanism for this association is immunologic stimulation through changes in the bowel flora, but Kummeling and associates¹⁰⁰ found no coincident increase in eczema or atopy, despite increased wheezing rates, which would argue against this mechanism. Greater antibiotic use might also represent a surrogate marker for a higher numbers of infections (perhaps viral) in early life.

Viral infections of the lower respiratory tract affect early childhood wheezing. Whether lower respiratory tract infection promotes sensitization to aeroallergens causing persistent asthma is controversial: childhood viral infections might be pathogenic in some children but protective in others.^102–106 Infants of mothers with allergy or asthma have a relatively persistent maturational defect in Th1 cytokine synthesis in the first year of life, which may play a role in the development of persistent or severe viral infections.¹⁰⁷ Severe viral infection of the lower respiratory tract in genetically susceptible infants who are already sensitized to inhalant allergens may lead to deviation toward Th2 responses promoting asthma. It is unclear whether these effects of lower respiratory tract infection are virus-specific (e.g., respiratory syncytial virus, rhinovirus) or whether synergistic exposures to allergens can induce asthma even in individuals who are not genetically susceptible. Interactions of genes with environmental exposures (including allergens, air pollution, environmental tobacco smoke and diet) modulate the host response to infections.^108,109 It remains controversial whether the occurrence or timing of childhood infection is pathogenic or protective for the development and long-term outcome of asthma and allergy and of nonallergic wheeze phenotypes. This controversy relates in part to small sample size, cross-sectional analysis, lack of precise case definition and incomplete microbial assessment in studies of this phenomenon.^110,111

Respiratory infections in early childhood are associated with early wheezing,¹⁰⁹ but it is unclear whether infection alone has a role in the development of persistent asthma. Repeated lower respiratory tract infection may affect infants who are already at risk for asthma because of family history or atopy.^63,112 Severe infection with certain viruses such as respiratory syncytial virus¹⁰⁶ and rhinovirus¹¹³ may play a role in persistent wheezing, although other studies have suggested no effect.¹¹⁴ Considered as a proxy for viral infections, daycare attendance is associated with greater incidence of early wheeze but lower incidence of persistent wheeze.¹¹⁵

Allergic sensitization

Total serum immunoglobulin E level, a surrogate for allergen sensitivity, has been associated with the incidence of asthma.¹¹⁶ High levels of immunoglobulin E at birth were associated with greater incidence of both atopy^117–119 and aeroallergen sensitivity but not necessarily asthma. However, sensitization to aeroallergens, particularly house dust mite, cat and cockroach allergens, is well documented as being associated with asthma.

Immune responses in the developing infant and young child may affect the development of asthma. For example, impairment in interferon γ production at 3 months was associated with a greater risk of wheeze.¹¹⁵ Immaturity in neonatal immune responses may promote the persistence of the Th2 immune phenotype and development of atopy,¹²⁰ but an association with persistent asthma is as yet unproven. More recent work has focused on the role of the innate immune system in handling and presentation of antigens and suggests that polymorphisms in Toll-like receptors^121,122 may play a greater role than previously recognized in the development of the skewed immune responses associated with persistent asthma.

Exposure to environmental tobacco smoke

Postnatal exposure to environmental tobacco smoke, especially from maternal smoking, has been consistently associated with respiratory symptoms of wheezing.^22,26,56 Exposure to environmental tobacco smoke also consistently worsens asthma symptoms and is a risk factor for severe asthma.^123,124

Exposure to animals

Although several studies have demonstrated a lower risk of development of atopy and asthma with exposure to farm animals in early life, the findings of studies of the influence of exposure to domestic cats and dogs have been inconsistent.^125,126 In some studies, exposure to cats was associated with a greater risk of allergic sensitization,¹²⁷ whereas other studies showed a lower risk.^128,129 Exposure to dogs may be protective not only against the development of specific sensitization to dog allergen^127,128 but also against other sensitization (e.g., to house dust mites) and asthma. Other studies of exposure to dogs have suggested that protection against wheezing may be mediated by high levels of endotoxin.¹³⁰

Gene-by-environment interactions

The effects of gene-by-environment interactions in asthma are complex. In some cases the genes code for enzymes that detoxify inhaled agents (e.g., glutathione transferase genes and environmental pollution), whereas in other cases, the exposures may have a more direct effect on gene expression via epigenetic mechanisms, such as DNA methylation or histone modification. Epigenetic modification of DNA is believed to be responsible for the phenotypic differences that develop over time between monozygotic twins.¹³¹ It has been suggested that it is principally through epigenetic modification of DNA that lifestyle and chemical exposures affect susceptibility to diseases.¹³² Nutrition and diet (e.g., folic acid, vitamin B₁₂), smoking, exposure to microbial products, maternal stress and maternal care are potential factors influencing fetal genetic expression, and a further window for epigenetic modification in early life may allow environmental factors to modify a child’s genome with the potential to cause or prolong allergy and asthma. Further work is needed to verify and understand these risks.

Sex and gender

Sex affects the development of asthma in a time-dependent manner. Until age 13–14 years, the incidence and prevalence of asthma are greater among boys than among girls.^133–142 Studies through puberty^{139,143–155} have shown a greater incidence of asthma among adolescent and young adult females^{133–135,156,157} and a greater proportion of males with remission of asthma.^136–140 Before age 12, boys have more severe asthma than girls,¹⁴² with higher rates of admission to hospital.^158–165 In contrast, adult females have more severe asthma than males, with more hospital admissions,^161,166,167 slower improvement,¹²⁰ longer hospital stays¹⁶¹ and higher rates of readmission.¹⁶⁸ Most authors have attributed these changes in prevalence and severity to events of puberty,^140,141 although mechanisms for differences between the sexes have not been established.

In childhood, airway hyperresponsiveness is more common and more severe among males;¹⁶⁹ however, airway hyperresponsiveness increases in females during adolescence,^170,171 such that by adulthood it is both more common and more severe among adult women.^{154,155,172–174} Similar findings have been reported from studies of atopy, which is more common in males before age 13;¹⁷⁵ during adolescence, the rate of new-onset atopy is higher among females,^176,177 so that by young adulthood the prevalence of atopy is almost equal.

The influence of some environmental risk factors such as allergens may be modified by sex. In one study of adults, 18% of women with asthma, but only 2.3% of men with asthma, had normal results on common tests related to atopy (negative skin prick tests, immunoglobulin E < 100 IU/mL and eosinophilia < 5%),¹⁷⁸ which suggested different disease mechanisms between the sexes. Interactions have been found between maternal and paternal history of atopy, breastfeeding and sex of the child in terms of the risk of asthma and atopy.⁷¹ Finally, the influence of obesity on the development of asthma is greater among women than among men and has not been shown to be influenced by caloric intake or physical activity.^179,180 Some have suggested that the relation between obesity and asthma may be causal, given the consistency, temporal association and dose–response relationships reported in the epidemiologic literature, but the mechanisms remain to be elucidated.¹⁸¹

Occupational asthma

Asthma related to workplace exposures has been documented in many occupational settings. Commonly associated occupations and exposures include car painting (isocyanates), hairdressing (various chemicals), domestic and commercial cleaning (cleaning solutions), health care professions (latex) and baking (flour dust), among many others.¹⁸⁹

The relation between exposure to substances in the work-place and new-onset adult asthma was explored among 6837 participants with no previously reported asthma symptoms in phase I of the European Community Respiratory Health Study.¹⁸⁷ Exposure to substances known to cause occupational asthma was associated with a higher risk of asthma overall (relative risk [RR] 1.6, 95% confidence interval [CI] 1.1–2.3) and of asthma defined by airway hyperresponsiveness (RR 2.4, 95% CI 1.3–4.6). Of common occupations, nursing was associated with the highest risk of occupational asthma (RR 2.2, 95% CI 1.3–4.0, p = 0.007), whereas exposure to an acute inhalation event, such as fire, mixing of cleaning agents or a chemical spill, was associated with an even higher risk (RR 3.3, 95% CI 1.0–11.1, p = 0.05). The population attributable risk of occupational exposure for adult asthma in that study ranged from 10% to 25%.

Other risk factors for adult asthma

Smoking tobacco¹⁹⁰ or marijuana^191,192 may give rise to symptoms suggesting asthma, although symptoms of cough and sputum production, suggesting chronic bronchitis, are more common. As in childhood, the differential diagnosis should include other forms of airway inflammation and other causes of intermittent dyspnea and wheezing, such as cardiac failure. However, new-onset asthma can occur at any age, without prior illness or concomitant disease. Atopy as a risk factor for asthma is less common with increasing age,¹⁹³ but occasionally it is the dominant trigger. Air pollution may affect adult asthma, but more often it is a factor worsening pre-existing asthma rather than a cause of incident asthma.^194–196