Pediatric Asthma Treatment Could Increase Risk of Diabetes, Other Conditions

Daniel Horton, MD

A new study from a Rutgers University-led team suggests children taking oral steroids may be at an increased risk for a multitude of health issues later in life, including diabetes, hypertension, and blood clots.

While investigators note the absolute risk for most children is low, results from the retrospective cohort analysis of more than 900k children indicate children receiving oral glucocorticoids to treat asthma or autoimmune diseases were at increased risk of cardiometabolic complications later in life.

“This study allowed us to put numbers on the association between oral steroids and rare, but potentially serious, complications in children,” said study investigator Daniel Horton, MD, MSCE, an assistant professor of pediatrics and epidemiology at Rutgers Robert Wood Johnson Medical School, in a statement. “While children receiving high-dose steroids were at substantially higher risk for developing diabetes, high blood pressure or blood clots relative to children not taking these medicines, the absolute risks of these complications were still small. The vast majority of children taking brief courses of steroids for conditions such as asthma, for instance, will not experience these complications.”

While incidence of cardiometabolic complications with oral steroids in older adults has been thoroughly researched, Horton and a team of colleagues sought to close the clinical knowledge gap related to this relationship in children—who investigators point out require different treatment regimens with oral steroids and have lower levels of baseline risk compared to their older counterparts. With this in mind, investigators designed their study as a retrospective cohort analysis using Medicaid Analytic eXtract (MAX) files from 2000-2010.

From MAX files, investigators identified a cohort of 932,517 eligible individuals for inclusion in their study. For inclusion in the study, participants needed to be between the ages of 1-18 years of age with 9 months or more of observable time, and 1 of 4 of the following disease diagnoses: inflammatory bowel disease (IBD), juvenile idiopathic arthritis (JIA), psoriasis, or attention-deficit/hyperactivity disorder (ADHD). ADHD was chosen to serve as a common non-immune reference condition for the purpose of comparison.

Children were excluded from the study if they had prior malignancies, chemotherapy, organ or hematopoietic transplantation, human immunodeficiency virus, or a non-JIA rheumatic disease. Investigators noted children diagnosed before age 1 were excluded due to questions about diagnostic validity.

For the purpose of analysis, glucocorticoid exposure, which was determined through pharmacy claims, was categorized as low (less than 0.25 mg/kg/day), medium (0.25-0.99 mg/kg/day), or high (1 mg/kg/day or more). Of the 932,517 children eligible for inclusion, 198,885 (21.3%) were considered glucocorticoid-exposed. The primary outcomes chosen for the study were incident treatment of diabetes mellitus, hypertension, or venous thromboembolism (VTE).

Results of the investigators’ analyses indicated crude rates were lowest for VTE with rates of 0.5 per million person-days (95% CI, 0.4-0.6) among unexposed patients versus 15.6 per million person-days (95% CI, 11.8-20.1) among exposed patients and highest for hypertension with rates of 6.7 per million person-days (95% CI, 6.5-7.0) among unexposed patients and 74.4 per million person-days (95% CI, 65.7-83.9) among exposed patients. The corresponding rates for diabetes mellitus were 5.3 per million person-days (95% CI, 5.1-5.5) among unexposed patients and 24.5 per million person-days (95% CI, 19.7-30.0) among exposed patients.

Investigates pointed out the absolute rates for all outcomes were higher in unexposed and exposed children with autoimmune diseases compared to those with ADHD.

Further analysis indicated strong dose-dependent relationships existed between current glucocorticoid exposure and all of the study outcomes. Specifically, the adjusted hazard ratios for high-dose glucocorticoids and incidence of outcomes were 5.93 (95% CI, 3.94-8.91) for diabetes mellitus, 19.13 (95% CI, 15.43-23.73) for hypertension, and 16.16 (95% CI, 8.94-29.22) for VTE. Investigators purport these results suggest strong relative risks but low absolute risks of new VTE, diabetes, and hypertension among children taking high-dose oral glucocorticoids.

This study, “Oral Glucocorticoids and Incident Treatment of Diabetes Mellitus, Hypertension, and Venous Thromboembolism in Children,” was published in the Journal of Epidemiology.