Corticosteroid Use May Modestly Reduce Mortality Risk in Sepsis
Our panel makes a weak recommendation for the administration of corticosteroids to people with all types and severity of sepsis based on new evidence.
The use of corticosteroids in sepsis appears to modestly reduce the risk for death and modestly increase neuromuscular weakness but the evidence is not definitive, according to a “Rapid Recommendation” published in the BMJ. Overall, this guideline makes a weak recommendation for corticosteroid use in sepsis as both steroids and no steroids are reasonable management options.
Broad spectrum antibiotics are the mainstay of sepsis treatment and supportive treatment may include vasoactive drugs and mechanical ventilation. Most guidelines currently do not advise the use of corticosteroids, except for refractory shock. However, 2 new studies with conflicting results have been published. This rapid recommendation was initiated because these 2 trials reached different conclusions regarding the use of steroids in sepsis. The study authors noted that these “results might change practice.”
In the ADRENAL study, which included 3658 patients with septic shock, there was no statistically significant difference in 90-day mortality in patients who received hydrocortisone vs patients given placebo. However, the APROCCHSS had incongruent findings. The clinical trial included 1241 patients with septic shock who received hydrocortisone plus fludrocortisone, which reduced 90-day mortality. Both of these studies were incorporated into a linked systematic review comparing corticosteroids with placebo.
“Our panel make[s] a weak recommendation [for the administration of] corticosteroids to people with all types and severity of sepsis, based on new evidence,” wrote the investigators. “Because we are not certain that they are beneficial, it is also reasonable not to prescribe them.”
Lamontagne F, Rochwerg B, Lytvyn L, et al. Corticosteroid therapy for sepsis: a clinical practice guideline [published online August 10, 2018] BMJ. doi:10.1136/bmj.k3284