Objective　To explore the efficacy and safety of colostrum oral smear or drip in the prevention of necrotizing enterocolitis in premature infants. Methods　Systematic retrieval of PubMed, Ovid-Embase, Web of Science, Cochrane Library, CNKI, Wan Fang database and VIP database was conducted to collect the randomized controlled trials (RCTs) of colostrum oral smear/drip in premature neonates. The retrieval period was from the establishment of the database to January 2019. Two reviewers independently screened the literature, extracted the data and evaluated the quality of the literature. Meta-analysis was performed using RevMan5.3 software. Results　Seven RCTs were included (368 subjects) totally. Meta-analysis showed that there were no statistical differences in incidence of necrotizing enterocolitis (OR=0.85, 95% CI: 0.40~1.79, P=0.67), length of stay (WMD=－7.09, 95% CI:－16.84~2.67, P=0.15), mortality (OR=0.64, 95% CI: 0.20~2.06, P=0.46) between colostrum oral smear/drip group and control group, while significant differences were found in the time to reach full enteral feeding (WMD= －1.90, 95% CI: －2.15~1.65, P<0.01) and days of antibiotic use (WMD= －2.57, 95% CI: －4.13~1.01, P<0.01). Conclusions　Oral smear/drip of colostrum could not reduce the incidence and mortality of necrotizing enterocolitis in premature infants, and could not shorten the hospitalization days of premature infants, but it could shorten the time to reach full enteral feeding and the days of antibiotic use in hospitalized premature infants.