Bottom line: In patients with meningitis, evidence shows that steroids given prior to or at the same time as antibiotics are effective at reducing adverse neurologic effects. However, there is insufficient evidence to indicate whether or not steroids can reduce lasting neurologic sequalae in patients with bacterial meningitis who have already received antibiotics.
Summary: Infectious Disease Society of America (IDSA) guidelines recommend administering steroids prior to antibiotics, but do not recommend administering steroids in patients who have already received antibiotic therapy because there is insufficient evidence that it improves outcomes. (See Bacterial Meningitis in Adults > Treatement > Steroids. In: DynaMed.)
Practice Guidelines for the Management of Bacterial Meningitis. For adults, there were 5 trials of adjunctive steroids in patients with bacterial meningitis at the time of the guidelines (2004), 3 RCT placebo-controlled, 1 randomized, no placebo control, 1 cohort study. In 4 of the 5 studies, results were inconclusive (steroids administered at different times among the studies.) In 1 RCT, patients randomized to receive dexamethasone for 2-4 days starting no more than 20 minutes prior to antibiotics, or at the same time as antibiotics. RESULTS: After 8 weeks, percentage of patients with an unfavorable outcome (15% vs. 25%; P = .03) and death (7% vs. 15%; P = .04) favored patients in the dexamethasone group.
Lancet Neurol. 2010 Mar;9(3):254-63
More recently, a meta-analysis of patient data on all double blind, placebo-controlled, RCTs evaluating use of dexamethasone in patients with meningitis. RESULTS: Data from 2,029 (1,019 receiving dexamethasone) patients in 5 RCTs was reviewed.
Outcomes at time of first follow-up for dexamethasone v. placebo
Death: 26.5% vs 27.2%; OR 0.97 (95% CI 0.79-1.19)
Death or severe neurological sequelae or bilateral severe deafness: 42.3% vs 44.3%; 0.92 (0.76-1.11)
Death or any neurological sequelae or any hearing loss: 54.2% vs 57.4%; 0.89 (0.74-1.07)
Death or severe bilateral hearing loss: 36.4% vs 38.9%; OR=0.89 (0.73-1.69)
Hearing loss only: 24.1% vs 29.5%; OR=0.77 (0.60-0.99, p=0.04)
Only significant difference is in reducing hearing loss among survivors.