Association of early life factors with prematurity-associated lung disease: prospective cohort study

Introduction Although bronchopulmonary dysplasia (BPD) is associated with lung function deficits in childhood, many who develop BPD have normal lung function in childhood, and many without BPD including those born at 33–34 weeks’ gestation, have lung dysfunction in childhood. Since the predictability of BPD for future lung deficits is increasingly doubted, we prospectively recruited preterm-born children to identify early life factors which are associated with lung function deficits after preterm-birth. Methods From 767 children aged 7–12 years, who had their respiratory symptoms assessed, and had spirometry before and after a bronchodilator in our Respiratory Health Outcomes in Neonates (RHiNO) study, 739 (544 preterm-born at ≤34 weeks’ gestation and 195 term-born) had satisfactory lung function. Data were analysed using multivariable logistic regression and mediation. Results When preterm-born children were classified according to their lung function, low lung function (prematurity-associated lung disease, PLD) was associated with BPD, gestation and intrauterine growth restriction on univariable logistic regression analyses. However, on multivariable logistic regression analyses, gestation (Beta=−0.153, se: 0.051, p=0.003) and intrauterine growth restriction (odds ratio 1.783, 95% confidence interval: 1.06, 3.00, p=0.029) remained significantly associated with later deficits of lung function but BPD (0.99; 0.52, 1.89, p=0.974) did not. Mediation analyses confirmed these results. Conclusions Although traditionally BPD has been associated with low lung function in later life, these data show that gestation and IUGR are significantly associated with PLD in childhood but BPD is not. By identifying children with PLD, we can better understand the underlying mechanisms and develop optimal therapies