BACKGROUND: Overweight and obesity are related to maternal and infant physical health, such as gestational diabetes, preeclampsia, and macrosomia. The purpose of this meta-analysis was to assess the effect of physical exercise on maternal and infant outcomes in overweight and obese pregnant women. METHODS: Two researchers independently searched Cochrane Library, Embase, PubMed, Web of Science, and ClinicalTrials.gov. for English-language articles based on randomized controlled trials examining physical exercise in overweight and obese pregnant women and its effect on maternal and infant outcomes. Primary outcomes were gestational weight gain and a relative risk of gestational diabetes. Secondary outcomes were gestational hypertension, preeclampsia, cesarean delivery, birthweight, large for gestational age, small for gestational age, macrosomia, and preterm birth. Risk bias was evaluated by Cochrane Collaboration's tool. The results of integration were reported as relative risks (RR), mean difference, or standard mean difference with 95% confidence intervals (CI). This meta-analysis was registered on PROSPERO on November 18, 2017, with registration number CRD42017081565. RESULTS: Thirteen studies involving 1439 participants were included. Physical exercise reduced gestational weight gain (mean difference = -1.14 kg, 95% CI = [-1.67 to -0.62], P < 0.0001) and the risk of gestational diabetes (RR = 0.71, 95% CI = [0.57-0.89], P = 0.004) in overweight and obese pregnant women. There were no significant differences in other outcomes such as gestational hypertension, preeclampsia, cesarean delivery, birthweight, large for gestational age, small for gestational age, macrosomia, and preterm birth. CONCLUSIONS: Prenatal exercise interventions reduced gestational weight gain and the risk of gestational diabetes for overweight and obese pregnant women, which reinforced the benefits of exercise during pregnancy. However, no evidence was found with respect to benefits and/or harm for infants. Consideration should be taken when interpreting these findings as a result of the relative small sample size in this meta-analysis. Further larger well-designed randomized trials may be helpful to assess the short-term and long-term effects of prenatal exercise on maternal and infant outcomes.