Corticosteroids for Treating Sepsis in Children and Adults

Do systemic corticosteroids reduce 28-day mortality for hospitalized children and adults with sepsis?

Systemic corticosteroids probably reduce 28-day mortality for adults (18 years and older) hospitalized with sepsis. Furthermore, systemic corticosteroids may decrease length of hospitalization, probably reduce in-hospital mortality, and may reduce 90-day all-cause mortality for children and adults hospitalized with sepsis. However, they have little to no effect on long-term (ie, 90 days to 1 year) all-cause mortality.¹ (Strength of Recommendation: B, inconsistent or limited-quality patient-oriented evidence.)

Sepsis, broadly defined as a dysregulated immune response to an infection, is a common pathologic syndrome in hospitalized children and adults. Each year, 1.7 million adults and up to 72,000 children develop sepsis in the United States.^2,3 The in-hospital mortality rate associated with sepsis is high at an estimated 17% to 33%.^4,5

This Cochrane review included 87 randomized controlled trials with 24,336 participants who had sepsis between 1976 and 2024 across 5 continents.¹ Sepsis was defined by the Sepsis 3 criteria or the presence of an infectious source with two symptoms of systemic inflammatory response syndrome and a sign of organ dysfunction. Of these trials, six included only children (excluding neonates), two included children and adults, and 79 included adults only. Steroid administration was compared with placebo or usual care. Any systemic corticosteroid was considered in the intervention groups (eg, cortisone, fludrocortisone, hydrocortisone, methylprednisolone, betamethasone, dexamethasone). This included continuous or bolus dosing of all dose ranges and treatment durations.

The primary outcome for this review was 28-day mortality. Secondary outcomes included length of hospitalization, long-term (ie, 90 days to 1 year) mortality, in-hospital mortality, and incidence of adverse effects (eg, superinfection, hyperglycemia, hypernatremia, neuropsychiatric effects [eg, delirium, confusion], muscle weakness, cardiac events, gastroduodenal bleeding).¹

Systemic corticosteroid administration probably improves 28-day mortality in children and adults compared with usual care (risk ratio [RR] = 0.89; 95% CI, 0.84–0.95; 72 trials; n = 22,915; number needed to treat = 36; moderate-certainty evidence). Subgroup analysis further demonstrated that systemic corticosteroids seem to improve 28-day mortality in adults (RR = 0.89; 95% CI, 0.83–0.95; 62 trials; n = 21,851) but do not have a significant effect on 28-day mortality in children.¹

In children and adults, systemic corticosteroids may reduce length of hospitalization (mean difference = −1.09 days; 95% CI, −1.85 to −0.34; 31 trials; n = 16,954; low-certainty evidence), may decrease intensive care unit length of stay (mean difference = −0.86 days; 95% CI, −1.67 to −0.05; 25 trials; n = 8,069; low-certainty evidence), probably decrease in-hospital mortality (RR = 0.9; 95% CI, 0.84–0.97; 40 trials; n = 17,459; moderate-certainty evidence), and may decrease 90-day all-cause mortality (RR = 0.89; 95% CI, 0.82–0.97; 13 studies; n = 8,360; low-certainty evidence). Corticosteroid use in patients with sepsis had no significant effect on long-term mortality.¹

Regarding adverse effects, systemic corticosteroids compared with placebo or usual care appear to increase the risk of hyperglycemia and hypernatremia.¹ These terms were not well defined, and the data were presented heterogeneously; therefore, the clinical implications of these findings are unclear. Compared to placebo or usual care, corticosteroids have little to no effect on the incidence of superinfection, gastroduodenal bleeding, or neuropsychiatric events, and it is unclear if they cause any increased risk of muscle weakness, cardiac events, or stroke. Because subgroup analyses distinguishing children vs adults were not reported in the meta-analysis, it is unclear whether these conclusions are true for each subgroup. Subgroup analyses distinguishing corticosteroid type, dose, duration, or mode of delivery did not show any differences in 28-day mortality among groups.

The results of this meta-analysis are consistent with a recent meta-analysis that also showed a possible short-term mortality benefit from systemic corticosteroids for patients with septic shock and a 2013 meta-analysis that showed no improvement in mortality in children with sepsis who were treated with systemic corticosteroids.^6,7 The Society of Critical Care Medicine recommends that patients with septic shock, defined as sepsis with hypotension unresponsive to intravenous fluids requiring vasopressors, receive high-dose intravenous corticosteroids (eg, hydrocortisone, 400 mg/day) for up to 3 days, but it does not make a recommendation regarding sepsis without shock.⁸ Further study is necessary, particularly regarding the use of systemic corticosteroids in children with sepsis.