BackgroundCorticosteroids are frequently used in practice to treat patients with neurological disorders. However, its effect for stroke and traumatic brain injury (TBI) remains controversial. This study aimed to systematically review and evaluate efficacy and safety of corticosteroids for the treatment of stroke and TBI.MethodsWe searched Ovid-Medline and Ovid-Embase databases for randomised controlled trials (RCTs) and cohort studies evaluating the efficacy and safety of corticosteroids in patients with ischaemic stroke, intracerebral haemorrhage (ICH), subarachnoid haemorrhage (SAH) or TBI. The treatment intervention was corticosteroid, and the control was placebo or routine care. Outcome measures were death, functional outcomes and adverse events. We calculated odds ratio (OR) and 95% confidence interval (CI) for the effect size, pooled the results using random-effects modelling, and assessed heterogeneity by I2 statistic.ResultsWe identified 47 studies (41 RCTs and 6 cohort studies). Nine studies enrolled patients with ischaemic stroke (n = 2806), 6 studies for ICH (n = 1229), 1 study recruited both ischaemic stroke (n = 13) and ICH (n = 27), 10 studies for SAH (n = 1318) and 21 studies for TBI (n = 12,414). Dexamethasone was the most used corticosteroid (28 studies). Corticosteroids reduced risk of death at 3 months after ischaemic stroke (n = 1791; 31% vs. 26%, OR 0.77, 95% CI 0.62-0.95; df = 1, I2 = 0%) and after ICH (1 study; n = 850; 44% vs. 27%, OR 0.48, 95% CI 0.35-0.64), had no effect on death at 1 month after SAH (1 study; n = 140; 22% vs. 32%, OR 1.73, 95% CI 0.81-3.68), and increased risk of death at 6 months after TBI (n = 10,755; 23% vs. 27%, OR 1.20, 95% CI 1.10-1.32; df = 6, I2 = 0%). The pooled analyses found no significant effect of corticosteroids on functional outcome after ischaemic stroke, ICH, SAH or TBI, respectively.ConclusionCorticosteroids reduced the risk of death and in selected patients with stroke, such as those with large artery occlusion after thrombectomy, but increased the risk of death after TBI, had no effect on functional outcomes. Further trials are needed to identify individual stroke patients who may benefit from corticosteroids.Systematic review registrationInternational Prospective Register of Systematic Reviews (CRD42023474473).