Hydrocortisone Does Not Reduce 90-Day Mortality in Septic Shock

WEDNESDAY, May 31, 2023 (HealthDay News) -- For patients with septic shock, intravenous hydrocortisone does not reduce 90-day mortality when compared with placebo, according to a study published online May 22 in NEJM Evidence.

Romain Pirracchio, M.D., M.P.H., Ph.D., from the University of California San Francisco, and colleagues pooled individual patient data from trials of septic shock examining the adjunctive use of intravenous hydrocortisone. The primary outcome was 90-day all-cause mortality, which was examined across predefined subgroups of patients.

Data were included from 24 eligible trials, of which 17 provided individual patient data (7,882 patients), and seven (5,929 patients) assessed 90-day mortality. The researchers found that for hydrocortisone versus placebo, the marginal relative risk (RR) for 90-day mortality was 0.93 (95 percent confidence interval [CI], 0.82 to 1.04; moderate certainty); RRs were 0.86 (95 percent CI, 0.79 to 0.92) and 0.96 (95 percent CI, 0.82 to 1.12) for hydrocortisone with and without fludrocortisone, respectively. No significant differential treatment effect was observed across subgroups. Little or no difference was seen in the secondary outcomes in association with hydrocortisone except for vasopressor-free days (mean difference, 1.24 days; 95 percent CI, 0.74 to 1.73; high certainty). Based on low-certainty evidence, hydrocortisone may be associated with an increased risk for hypernatremia and muscle weakness (RRs [95 percent CIs], 2.01 [1.56 to 2.60] and 1.73 [1.49 to 1.99], respectively).

"This study shows that if the effect of hydrocortisone on the mortality of septic shock is modest, this treatment makes it possible to spare the exposure of patients to vasopressor drugs and to prevent their complications," Pirracchio said in a statement. "The combination of fludrocortisone with hydrocortisone seems to provide a greater benefit in terms of survival."

Several authors disclosed financial ties to industry.

Abstract/Full Text