ObjectiveIntravenous (IV) racemic ketamine and intranasal (IN) esketamine have demonstrated rapid antidepressant effects in treatment-resistant depression (TRD). This systematic review aims to evaluate the efficacy and safety of ketamine and esketamine at various dosages for depression. MethodsWe included randomized controlled trials (RCTs) with parallel group dose comparison of ketamine and esketamine for depression/TRD. Ovid Medline, Embase, PsycINFO, Scopus and Cochrane databases were searched. Standardized mean differences were calculated using Hedges'-g to complete random effects meta-analysis. The efficacy outcomes were changes in depression outcomes for IV ketamine and IN esketamine respectively. Safety was assessed by reported adverse effects. ResultsA random effects meta-analysis of studies (n = 12) showed efficacy in reducing depression symptoms with IV ketamine (Hedges'g = 1.52 [0.98–2.22], Z = 4.23, p < 0.001) and IN esketamine (Hedges' g = 0.31 [0.18–0.44], Z = 4.53, P < 0.001) compared to control/placebo. Treatment response was observed at IV ketamine doses ≤0.2 mg/kg, >0.2–0.5 mg/kg and > 0.5 mg/kg. Higher IV ketamine doses (>0.5 mg/kg) did not lead to greater treatment response. Esketamine doses of 56–84 mg were superior to 28 mg dose. LimitationsOverall quality of evidence was low and limited by small number of studies. Publication bias was high. ConclusionsThis meta-analysis suggests that IV ketamine may be efficacious at doses as low as 0.2 mg/kg, with increasing dose response at 0.5 mg/kg, without demonstrable increased benefit at 1 mg/kg, based on a small number of studies. Efficacy for IN esketamine increases with doses above 28 mg with best response being found between 56 and 84 mg for reducing depressive symptoms.