ObjectiveTo evaluate the effectiveness of levetiracetam (LEV) added on to usual care, in treating children refractory partial seizure epilepsy.MethodsWe searched the Cochrane library, EMBASE and PubMed between January 1998-January 2017, We systematically searched CNKI database and Wanfang data, Chinese biology medline and the manual retrieval related magazines.RevMan 5.3 statistical software for Meta analysis.ResultsAccording to the enrollment criteria, fourtrials were included involving 498 participants according to the intent-to-treat, 268 for LEV, and 230 for placebo groups.We assessed the following outcomes: 50% or greater seizure reduction, seizure freedom, adverse effects, proportion of dropouts and quality of life. There was no evidence of statistical heterogeneity between trials.We assessed outcomes by using a meta-analysis to calculate odds ratio (OR) with 95% confidence intervals (95% CI). For the 50% or greater reduction in focal seizure frequency outcome, the OR was significantly in favour of LEV [OR=2.94, 95% CI(1.99, 4.34)].Participants were significantly more likely in LEV groups than placebo groups to get seizure free[OR=5.31, 95% CI(2.49, 11.32)]. There was no significance between LEV groups and placebo groupsin the rate of Treatment withdrawal[OR=0.76, 95% CI(1.32, 1.82)]. Somnolence[OR=2.57, 95% CI(1.36, 4.86)]and changes in behaviour [OR=2.54, 95% CI(1.56, 4.14)] were significantly associated with LEV. Other adverse effects were not significantly associated with LEV in children.ConclusionThe existing evidence suggests that LEV add in treatment of children refractory epilepsy have definite curative effect, LEV long-term treatment effect is stable, good security, retention rate is higher, can be used in clinical further promotion.