Corticosteroids as Anti-inflammatory Therapy in Sepsis
According to the literature, the use of glucocorticoids for intensive care patients, especially for those with sepsis, is controversial. Is there a real benefit for all patients or should utilization of these medications be limited to selected patients?

Carlos Alexandre Ferreira de Oliveira, MD

You pose an interesting and controversial question. There is tremendous variability in how physicians use corticosteroids in the intensive care unit. Their use has been investigated over many decades with few true indications, aside from absolute adrenal insufficiency.

For patients with severe sepsis or septic shock, many trials have used corticosteroids as anti-inflammatory therapy to interrupt the pathophysiologic cascade of events. These landmark trials generally failed to find any significant benefit to corticosteroid therapy for patients with severe sepsis or septic shock. These studies have generally been small and perhaps underpowered for determining major differences in survival. Combining these clinical trials by meta-analysis provides a useful additional measure of efficacy of corticosteroids. No benefit was found and perhaps some harm from corticosteroid administration (relative risk = 1.10).

Recent investigations have focused on longer duration therapy with lower doses of corticosteroids. Briegel and colleagues reported a shorter median duration of vasopressor requirement in septic shock patients treated with hydrocortisone, though there was no statistical difference in shock reversal or mortality. Bollaert and coworkers randomized septic shock patients with vasopressor requirements greater than 48 hours to receive hydrocortisone or placebo. Treated patients in this study had significantly more rapid reversal of shock and reduced mortality at 28 days.

It has been hypothesized that the benefit of corticosteroids may occur with moderately supraphysiologic doses and that corticosteroids may be most useful for patients with inadequate adrenal reserve. Annane and colleagues identified septic shock patients with differential risk for death based upon corticotropin stimulation testing. Using this classification, they randomized septic shock patients to receive hydrocortisone (50 mg every 6 hours) with fludrocortisone (50 mcg every day) or 2 placebos. Patients with relative adrenal insufficiency treated with steroids were noted to more frequently resolve vasopressor requirements and experience improved survival, while patients with adequate adrenal reserve did not benefit from corticosteroid administration. Application of these results to clinical practice has been variable, and remains difficult because of the selected population for whom corticosteroids seem to benefit as well as the logistic issues of rapidly identifying patients with relative adrenal insufficiency.