Background: Risk of death may influence the efficacy of anti-inflammatory agents in sepsis. "Physiologic" dose corticosteroids, while improving survival in earlier trials with higher control mortality rates (>50%), were not beneficial in the recent CORTICUS trial with lower control mortality (31%). We investigated whether risk of death altered the effects of hydrocortisone in a mouse pneumonia model.
Methods: Mice (n=637) challenged with high, medium or low intratracheal E. coli doses were randomized to receive one of three hydrocortisone doses (5, 25 or 125 mg/kg) or normal saline (NS) only (control) for 4 days. All animals were treated with similar volumes of ceftriaxone and NS support following E. coli and were observed for 168 h.
Results: Decreasing E. coli doses reduced control mortality rates (from 94 to 12%). In similar patterns (not significant) each hydrocortisone dose increased the odds ratio (OR) of survival (95% confidence interval) with each E. coli dose (ORs ranging from 1.2 [0.4, 3.7] to 6.1 [0.6, 61.0]). The effect of hydrocortisone on the OR was not related to control mortality rate (r=-0.13, p=0.29) and overall was highly significant (2.04 [1.37, 3.03], p=0.0004). In randomly selected animals 48 h after the highest E. coli dose, compared with the control, hydrocortisone (125 mg/kg) significantly decreased IL-6, INFgamma, and nitric oxide levels.
Conclusions: In this mouse model the beneficial effects of hydrocortisone were independent of risk of death. These findings suggest that factors other than risk of death may underlie the differing effects of corticosteroids in recent sepsis trials.