Early life origins of asthma : genetic and environmental factors in twin and kin

Abstract

During the last decades there has been an upsurge of studies investigating how early life exposures may affect subsequent health outcomes in childhood. For instance, low birth weight and exposure to antibiotics during fetal or early life have been suggested to increase the risk of childhood asthma. Twin- and sibling comparisons can help to account for confounding factors shared within families and to shed light on potential causal pathways.

Thus, in Study I, we aimed to investigate if low birth weight, as a proxy for fetal growth restriction in twins, was associated with the development of asthma in a cohort of 10 918 Swedish twins aged 9 or 12 years. We found a significant association which was thought to be explained by unique individual factors and not due to familial confounding or gestational age. It can be speculated that restricted fetal growth affects lung development in utero which influences the risk of developing asthma in childhood.

In Study II, we aimed to further understand the adverse effects of restricted fetal growth, by specifically investigating its association with childhood lung function in a cohort of 576 twins aged 9-14 years at invitation to the study. We found a significant association between fetal growth restriction and reduced forced expiratory volume in the first second, which could be explained by unique factors of each twin. Similar effects were found in non-asthmatic individuals, whereas other factors may be of importance for the association between fetal growth and lung function in individuals with asthma.

To be able to study asthma in register-based studies, a valid measure of the disease is needed. In Study III, medical records for roughly 1700 individuals, corresponding to prescription dates of asthma medications or to discharge dates accompanying asthma diagnoses and identified from population-based drug- and patient registers, were collected from health care units and evaluated against pre-defined criteria of asthma. We found a high positive predictive value for asthma medication as a proxy for asthma in older children and the majority of children with an asthma diagnosis in the patient register fulfilled pre-defined criteria of asthma.

In Study IV, the previously suggested association between antibiotic exposure in fetal and early life and childhood asthma (based on the validated outcome measure from Study III) was investigated in a cohort of 493 785 children. We found an association between antibiotic exposure both in fetal and early life and asthma. Yet, sibling control analyses suggested that the associations were due to shared factors within families, and confounding by indication or reverse causation due to respiratory infections.

In conclusion, shared genetic and environmental factors contributed to the association between antibiotics and asthma, but not between fetal growth and asthma and lung function, respectively. Genetically informed designs to control for familial confounding are useful tools to help provide a clearer understanding of the etiology of asthma. In addition, asthma identified from population-based registers can be used as a validated outcome measure and contribute towards future studies on asthma using register-based data.