Respiratory syncytial virus infection during infancy and asthma during childhood in the USA (INSPIRE): a population-based, prospective birth cohort study

Abstract

Background: Early-life severe respiratory syncytial virus (RSV) infection has been associated with the onset of childhood wheezing illnesses. However, the relationship between RSV infection during infancy and the development of childhood asthma is unclear. We aimed to assess the association between RSV infection during infancy and childhood asthma.

Methods: INSPIRE is a large, population-based, birth cohort of healthy infants with non-low birthweight born at term between June and December, 2012, or between June and December, 2013. Infants were recruited from 11 paediatric practices across middle Tennessee, USA. We ascertained RSV infection status (no infection vs infection) in the first year of life using a combination of passive and active surveillance with viral identification through molecular and serological techniques. Children were then followed up prospectively for the primary outcome of 5-year current asthma, which we analysed in all participants who completed 5-year follow-up. Statistical models, which were done for children with available data, were adjusted for child's sex, race and ethnicity, any breastfeeding, day-care attendance during infancy, exposure to second-hand smoke in utero or during early infancy, and maternal asthma.

Findings: Of 1946 eligible children who were enrolled in the study, 1741 (89%) had available data to assess RSV infection status in the first year of life. The proportion of children with RSV infection during infancy was 944 (54%; 95% CI 52-57) of 1741 children. The proportion of children with 5-year current asthma was lower among those without RSV infection during infancy (91 [16%] of 587) than those with RSV infection during infancy (139 [21%] of 670; p=0·016). Not being infected with RSV during infancy was associated with a 26% lower risk of 5-year current asthma than being infected with RSV during infancy (adjusted RR 0·74, 95% CI 0·58-0·94, p=0·014). The estimated proportion of 5-year current asthma cases that could be prevented by avoiding RSV infection during infancy was 15% (95% CI 2·2-26·8).

Interpretation: Among healthy children born at term, not being infected with RSV in the first year of life was associated with a substantially reduced risk of developing childhood asthma. Our findings show an age-dependent association between RSV infection during infancy and childhood asthma. However, to definitively establish causality, the effect of interventions that prevent, delay, or decrease the severity of the initial RSV infection on childhood asthma will need to be studied.

Funding: US National Institutes of Health.

Figure 1:

Graphical representation showing the confounding effect of host genetics on the potential causal relationship between RSV infection in infancy and childhood asthma. (1A) Prior studies in this field have focused exclusively on severe RSV infection (usually defined as the presence vs. absence of RSV bronchiolitis requiring hospitalization) in infancy with childhood asthma (a) and have thus lacked a true control group of children without RSV infection in infancy. Furthermore, because both the severity of an RSV infection and the onset of childhood asthma may share a common genetic origin, or an RSV bronchiolitis requiring hospitalization in infancy may simply represent the first acute asthma exacerbation in an otherwise genetically predisposed child, these studies were highly susceptible to confounding by host genetics (b and c). (1B) In the current study, we examined the effect of being uninfected with RSV in infancy on the development of childhood asthma (d) and thus included a true control group of children not infected with RSV in the first year of life. In addition, because RSV is ubiquitous and nearly everyone is infected with it by age 2–3 years, being uninfected vs. infected with RSV in infancy is a natural event and less likely to be associated with host genetics (e). Hence, our study design is likely to mitigate confounding by shared genetic susceptibility. Definition of abbreviations: RSV = Respiratory syncytial virus.

Figure 2:

Flow diagram of the hierarchical algorithm used for the classification of RSV infection in infancy. The number (%) of eligible children enrolled in the study with RSV infection in infancy and 5-year current asthma are also shown. Definition of abbreviations: INSPIRE = Infant Susceptibility to Pulmonary Infections and Asthma Following RSV Exposure Study, RSV = Respiratory syncytial virus, RT-qPCR = Reverse transcription-quantitative polymerase chain reaction.

Figure 3:

The association of being uninfected with RSV in infancy with the primary and secondary outcomes. The bar plots show the % of children with 5-year current asthma (3A), recurrent wheeze (3B) at each of the measured timepoints, and 5-year current asthma inflammatory subtype ascertained using evidence of aeroallergen sensitization by skin prick testing or blood specific IgE testing at age 3 years (3C) in children infected and uninfected with RSV in infancy. The p-values shown for the comparison between groups were calculated using chi-squared tests. The number of children with each outcome and the total number of children for each group are shown inside the bars. Definition of abbreviations: RSV = Respiratory syncytial virus,

Figure 4:

Predicted probability of recurrent wheeze over the first 4 years of life by RSV infection in infancy. The solid lines represent the predicted probability for each group. The adjacent shaded bands represent the corresponding lower and upper 95% CIs. The estimates were obtained from a generalized estimating equation regression model with a Poisson random component and log link for repeated measures (using an independent working correlation) and correcting covariance matrices using the Huber-White robust sandwich method. The model included the child’s sex, race and ethnicity, ever breastfeeding, daycare attendance in infancy, exposure to secondhand smoking in utero or in early infancy, and maternal asthma as covariates. Definition of abbreviations: CI = Confidence interval, RSV = Respiratory syncytial virus.