Some of the most common practices in Emergency Medicine are only weakly tested or defined – including steroids for acute respiratory illness. What is the true minimum effective dose? How many days – 3, 5, 7, or 14? Burst or taper? Much of our practice is based on habit and mimicry, along with the general evidence that, despite ourselves, we don’t seem to be doing much harm.
This is the REDUCE trial, a multi-center, randomized, double-blind, non-inferiority comparison between a 5-day and a 14-day course of 40mg oral prednisone for acute COPD exacerbation. And…it found no difference in the primary outcome measure. So, then, all’s well.
Except, their outcome measure is utterly bizarre – re-exacerbation within six months? I cannot fathom how a 1-2 week period of steroids could have any causative association with outcomes more than a handful of half-lives after cessation of treatment. Perhaps they theorize the short-term steroid exposure is insufficient to avoid long-term damage secondary to the acute inflammation?
There are also some potentially confounding differences in baseline characteristics. There were 9% more smokers and 5% more home oxygen in the short-term treatment group – which could favor conventional treatment – but then 4% fewer were on daily steroids during the treatment period and 8% fewer received concurrent antibiotics in the conventional treatment group – which could favor the short-term treatment group. Some of these differences would have more effect on short-term outcomes, while others would affect long-term outcomes. I don’t know if there are true clues in the Kaplan-Meier curves they present – because the sample size is small enough these variations might just be occurring due to chance – but it appears there’s a possible hazard towards early re-exacerbation in the 5-day group, followed by regression towards equivalence by six months.
However, these issues aside, their conclusion is probably valid. Their predefined threshold for non-inferiority was 15% – and they easily cleared that bar. All the confounders are probably not of significant magnitude to affect the overall result at that threshold – even for shorter, more relevant follow-up time periods. Additionally, this is otherwise consistent with the other evidence that short-courses of steroids are absolutely acceptable in this context.
“Short-term vs Conventional Glucocorticoid Therapy in Acute Exacerbations of Chronic Obstructive Pulmonary Disease”
www.ncbi.nlm.nih.gov/pubmed/23695200