Obese or overweight children with asthma could be using inhaled corticosteroids (ICS) to no avail, combined results from observational studies suggest.
Dr. Cristina Longo
Using Mendelian randomization, a method for reducing bias in observational studies, investigators from the University of Amsterdam Medical Center performed an analysis of data from four cross-sectional studies and one cohort study on a total of 1,511 children with asthma.
They showed that every 1-unit increase in the body mass index (BMI) z score was associated with a more than twofold higher odds ratio for exacerbation, reported Cristina Longo, PhD, a former postdoctoral fellow at AMC, and assistant professor of medicine at the University of Montreal.
“In this large, multicenter Mendelian randomization study, our findings support current evidence that children with higher BMI status respond inadequately to inhaled corticosteroids, and that this association is likely not explained by measured confounding or reverse causation,” she said in an oral abstract presentation during the European Respiratory Society International Congress.
Unmeasured confounding
The obese-asthma phenotype in children is characterized by reduced lung function, high symptom expression, poor response to ICS, and high health care utilization.
“While most observational studies suggest that weight status is associated with asthma exacerbations, despite using inhaled corticosteroids, it’s unclear whether these associations may be due to unmeasured confounding or reverse causation, which captures the idea that perhaps obesity is a consequence rather than a cause of uncontrolled severe asthma,” she said.
Traditional observational studies of the obesity-asthma link rely on comparing data on asthma in a target population and comparing nonobese patients with obese patients. The problem with this method, Dr. Longo contended, is that the exposure assignment – weight status – is not random, and could lead to bias from potential imbalance of confounders, leading to unintentionally biased results.
In contrast, Mendelian randomization uses genetic data to approximate random assignment of exposures, using a risk score for BMI based on genetic susceptibility. The score is based on the accumulation of genetic variants (single-nucleotide polymorphisms, or SNPs) that predispose individuals to obesity, with higher numbers of variants results in a higher risk score.
The scores are then used to determine the comparison groups for evaluating the obesity-asthma association.
Alphabet soup
Dr. Longo and colleagues analyzed data on a total 1,511 children enrolled in four observational studies (PACMAN, PAGES, HPR, CLARA) and one cohort study (ALSPAC).
They included children with an asthma diagnosis who used ICS and had available information on both BMI and genetics.
The Mendelian randomization analysis was based on a weighted allele score based on 97 SNPs predictive of BMI based on large-scale genomewide association studies. The exposure for the analysis was age- and sex-adjusted BMI z scores based on World Health Organization growth charts for children.
They found that using the Mendelian randomization approach, for each standard deviation increase in BMI, the OR for any parent-reported asthma exacerbations, including urgent care visits or use of oral corticosteroids, was 2.31 (95% confidence interval, 1.26-4.25).
In contrast, if the traditional observational model had been used, the OR would be a nonsignificant 1.10 (95% CI, 0.99-1.22).
“Treatment guidelines recommend steroids for children with asthma who have a higher-than-normal BMI,” Dr. Longo said in a statement. “Our research group felt that the one-size fits-all approach to treating children with asthma with inhaled steroids as their first-line treatment, particularly those with excess weight, warrants revision. At the very least, research identifying potential alternative treatments should be encouraged and prioritized, especially since 30% of children with asthma are also obese. With the childhood obesity epidemic rising, we expect this percentage to increase meaning this problem of poor control will be seen more frequently in routine clinical practice.”
Christopher E. Brightling, PhD, professor of respiratory medicine at the University of Leicester (England), commented that “this is very good and fascinating research with findings that are important and novel.
“It sheds light on the complex interplay between genes, weight, and response to inhaled corticosteroids, underscoring the need to combine drug treatments with lifestyle and diet modifications. Policy makers, health care providers and families need to do much more to tackle the growing obesity epidemic in young people,” he said.
Dr. Brightling was not involved in the study.
The study was supported by the ERS and the European Union’s H2020 research and innovation program. Dr. Longo was a Horizon 2020 Marie-Sklodowska Cure Respire-3 fellow. Dr. Brightling reported no relevant disclosures.