Clinical Question: Are inhaled corticosteroids associated with fractures in patients with chronic obstructive pulmonary disease (COPD)?
Setting: Outpatient (any)
Study Design: Case-control
Synopsis: The authors used Veterans Affairs databases to identify a cohort of 40,517 patients with a new diagnosis of COPD, and then identify 1,708 patients from this group who had a fracture during an average of 1.75 years after the diagnosis. These patients were then compared with 6,817 COPD patients without fractures who were matched for age, sex, and date of diagnosis. Using the pharmacy database, the authors determined inhaled corticosteroid exposure from prescriptions dispensed between the dates of COPD diagnosis and data extraction. The authors converted the dosages to beclomethasone equivalents and then categorized the average daily dosages into low (less than 300 mcg per day), medium (300 to 699 mcg per day), and high (700 mcg per day or more).
Only 21 percent of the patients with fractures were exposed to inhaled corticosteroids, compared with 22 percent of control patients (unadjusted odds ratio [OR] = 0.96; 95 percent confidence interval [CI], 0.84 to 1.09). When comparing different dosage classifications, they found 6.3 percent of the patients with fractures used low-dosage inhaled corticosteroids compared with 7.7 percent of the control patients. Similarly, 8.7 percent of the patients with fractures used medium-dosage inhaled corticosteroids compared with 9.2 percent of the control patients. However, slightly more patients with fractures (6.3 percent) used high-dosage inhaled corticosteroids than did control patients (5.2 percent). After adjusting for comorbid conditions, medications, and hospitalizations, the only significant association with fracture risk was in patients who currently were using high-dosage inhaled corticosteroids (adjusted OR = 1.68; 95 percent CI, 1.10 to 2.57). It is unclear if this is a meaningful or chance association.
Bottom Line: In this study, only the current use of high dosages of inhaled corticosteroids was associated with an increase in fracture risk. It is unclear whether this is a chance association. Finally, this study cannot establish a causal relationship; users of high-dosage steroids may be more debilitated and more sedentary, thereby increasing the risk of fracture. (Level of Evidence: 2b)