The Association of Inhaled Corticosteroid Use with Serum Glucose Concentration in a Large Cohort

Background: Inhaled corticosteroids (ICSs) are widely used in the treatment of obstructive lung disease. ICSs have been shown to be systemically absorbed. The association between ICS and serum glucose concentration is unknown. Methods: To explore the association of ICS dosing with serum glucose concentration, we used a prospective cohort study of US veterans enrolled in 7 primary care clinics between December 1996 and May 2001 with 1 or more glucose measurements while at least 80% adherent to ICS dosing. The association between ICS dose from pharmacy records standardized to daily triamcinolone equivalents and serum glucose concentration was examined with generalized estimating equations controlling for confounders, including systemic corticosteroid use. Results: Of the 1698 subjects who met inclusion criteria, 19% had self-reported diabetes. The mean daily dose of ICS in triamcinolone equivalents was 621 μg (standard deviation 555) and 610 μg (standard deviation 553) for subjects with and without diabetes, respectively. After controlling for systemic corticosteroid use and other potential confounders, no association between ICS and serum glucose was found for subjects without diabetes. However, among subjects with self-reported diabetes, every additional 100 μg of ICS dose was associated with an increased glucose concentration of 1.82 mg/dL (P value .007; 95% confidence interval [CI], 0.49-3.15). Subjects prescribed antiglycemic medications had an increase in serum glucose of 2.65 mg/dL (P value .003; 95% CI, 0.88-4.43) for every additional 100 μg ICS dose. Conclusion: Among diabetic patients, ICS use is associated with an increased serum glucose concentration in a dose-response manner.