To quantify the effects of castration in male beef cattle on welfare indicators based on cortisol concentration, average daily gain (ADG) and vocalization, a systematic review and meta-analysis were performed.We searched on five electronic databases (CAB Abstracts, ISI Web of Science, PubMed, Agricola and Scopus) from January 1900 to May 2015 and included conference proceedings and electronically contacted experts, as well as we checked references of relevant review papers. Inclusion criteria were complete studies using beef cattle until one yearof age undergoing castration that analysed cortisol level, ADG or vocalization.Data were extracted using pre-defined protocols. The included documents were written in English, Spanish, Portuguese or Italian.Random effect meta-analyses were conducted for each indicator separately with the mean of control and treated group. Possible influences of study characteristics and quality were assessed in meta-regression analyses.A total of 18 prospective publications reporting 23 studies and 156 trials were included in the MA involving 1,617animals.Significant between studies heterogeneity was observed for MA results when analysing cortisol and ADG.Regardless the control group and the castration technique, the comparison analyses showed no changes (P ? 0.05) on cortisol levels when castration was performed without drug administration.We found no evidence (P ? 0.05) for multi-modal therapyin decrease cortisol concentration 30 min after surgical procedure.Anaesthesia did not decrease cortisol level (MD = 0.411 nmol/L; P = 0.077; 95% CI:-0.868, 0.045) 120 min after surgical castration compared to castrated group without drug administration.Random-effect meta-analysis suggested an increase in ADG in surgical (MD = 0.231 g/d; P = 0.010; 95% CI:0.056, 0.405) and non-surgical castration (MD = 0.883 g/d; P