OBJECTIVE：To systematically evaluate the safety of meropenem for neonatal infection ，and to provide evidence-based reference for safe use of it in the neonatal population. METHODS ：Retrieved from PubMed ，Embase，Cochrane Library，ISI Web of Science ，International Health Technology Assessment Network Website ，China Journal Full-text Database ， Wanfang Database ，CBM，Chinese Sci-tech Periodical Full-text Database ，randomized controlled trials （RCTs）about meropenem or meropenem combined other drugs （trial group ）versus the similar drugs that could replace meropenem （control group ）for neonatal infection were collected during the inception to May 1st，2021. After literature screening and data extraction ，the quality of included literatures were evaluated with Cochrane systematically evaluator manual 5.1.0. Meta-analysis was conducted with RevMan 5.3 software. RESULTS ：A total of 25 RCTs were included ，involving 2 090 children. Results of Meta-analysis showed that the incidence of overall ADR in trial group was significantly lower than control group [RR ＝0.53，95％CI（0.44，0.65），P＜0.000 01]. Results of subgroup analysis showed that the incidence of overall ADR in trial group was significantly lower than control group receiving ceftazidime [RR ＝0.55，95％CI（0.41，0.74），P＜0.000 1]，tigecycline [RR ＝0.37，95％CI（0.23，0.59），P＜0.000 1]， ceftriaxone [RR ＝0.53，95％CI（0.35，0.80），P＝0.003]. The incidence of overall ADR in trial group with neonatal purulent meningitis [RR ＝0.63，95％CI（0.44，0.92），P＝0.02]，severe neonatal multidrug-resistant bacterial infection [RR ＝0.37，95％CI（0.25， 0.55），P＜0.000 01]，neonatal severe bacterial infection [RR ＝ 0.67，95％CI（0.48，0.94），P＝0.02] were significantly lower than control group. The incidence of specific ADR such as mail： rash，gastrointestinal reaction ，hemoglobin reduction in trialgroup were significantly lower than control group （P＜0.05）. There was no statistical significance in the incidence of specific ADR between 2 groups，such as elevated transaminase ，secondary fungal infection and renal injury （P＞0.05）. Results of bias analysis showed that when the incidence of overall ADR was used as index ，there was a certain degree of publication bias in this study ，when the incidence of specific ADR was used as index ，there was less possibility of publication bias in this study. CONCLUSIONS：Meropenem is safe in the treatment of neonatal infection ，especially in the treatment of neonatal purulent meningitis，severe neonatal multidrug-resistant bacterial infection and neonatal severe bacterial infection ，it is superior to ceftazidime，tigecycline，ceftriaxone and other antibacterial drugs in safety.