Objective:To evaluate the effect of glucocorticoid therapy in children with sepsis.Methods:Databases including PubMed, EMbase, The Cochrane Library, Wanfang Data, VIP and CNKI were searched from inception to July 2019 to collect randomized controlled trials(RCTs)about glucocorticoid for children with sepsis.Two independent researchers screened the literature, extracted data, and assessed the risk of bias in the included studies.Data results were analyzed by RevMan5.3 software for Meta-analysis.Results:A total of 7 RCTs, involving 1 359 patients were included.The results of Meta-analysis showed that, compared with the control group, the corticosteroid treatment group did not improve mortality( OR=1.09, 95% CI 0.61, 1.97, P=0.76), length of stay( MD=-3.13, 95% CI -8.29, 2.03, P=0.23), PICU stay time( MD=-2.33, 95% CI -5.03, 0.37, P=0.09), mechanical ventilation time( MD=-2.01, 95% CI -4.66, 0.65, P=0.14), in hospital infection rate( OR=1.58, 95% CI 0.27, 9.21, P=0.61). Conclusion:The existing evidence shows that glucocorticoid adjuvant therapy can not effectively reduce the mortality, hospital stay, PICU stay time, mechanical ventilation time, hospital infection rate of children with sepsis, but it is limited by the number and quality of the included studies, which needs further large-scale, multicenter, double-blind, randomized controlled trials to verify.