In Reply: We agree (and acknowledged) that well-designed observational comparative effectiveness research studies, such as the one conducted by Dr Lindenauer and colleagues,1 cannot completely exclude the possibility of selection bias and confounding. As shown in their study, current practice strongly favors high-dose intravenous corticosteroids (median, 600 mg/d prednisone equivalent) compared with lower doses of oral corticosteroids (median, 60 mg/d prednisone equivalent) for patients hospitalized with COPD exacerbations, consistent with the belief that “more is better than less.” However, we are aware of little evidence to support this belief. Also, results of a modest-sized randomized clinical trial in hospitalized patients indicate that lower-dose systemic corticosteroids (eg, 30 mg/d prednisone equivalent) is efficacious compared with placebo.2 The potential for harm increases with higher doses of corticosteroids, including adverse effects on mood, sleep, glycemic control, and volume status. Compared with oral corticosteroids, intravenous corticosteroids add nursing costs, patient discomfort, and risk of thrombophlebitis.
Krishnan JA, Mularski RA. Corticosteroid Administration and Treatment Failure in Acute Exacerbation of Chronic Obstructive Pulmonary Disease—Reply. JAMA. 2010;304(14):1554-1556. doi:10.1001/jama.2010.1440