This study aimed to examine the effect of lifestyle intervention on the risk of gestational diabetes mellitus (GDM). We searched PubMed, Springer and other databases to retrieve articles published in English and Chinese up to 30 September 2015. The inclusion criteria were randomized controlled trials evaluating the effects of lifestyle intervention on risk of GDM. Exclusion criteria were studies with prepregnancy diabetes mellitus or interventions with nutrient supplements. Random-effect and fixed-effect model analyses were used to obtain pooled relative risks and 95% confidence intervals (CIs) of diet and physical activity on the risk of GDM. Subgroup analyses were performed to check the consistency of effect sizes across groups where appropriate. We identified 29 randomized controlled trials with 11,487 pregnant women, addressing the effect of lifestyle intervention on the risk of GDM. In the pooled analysis, either diet or physical activity resulted in an 18% (95%CI 5-30%) reduction in the risk of GDM (P = 0.0091). Subgroup analysis showed that such intervention was effective among women with intervention before the 15th gestational week (relative risk: 0.80, 95%CI 0.66-0.97), but not among women receiving the intervention afterwards. We conclude that lifestyle modification during pregnancy, especially before the 15th gestational week, can reduce the risk of GDM.