LAUSR

Objective To explore the effect of exercise during pregnancy on the maternal and neonatal health outcomes. Methods Eligible papers were systematically retrieved from PubMed, Embase, OVID, and ScienceDirect. Two researchers independently extracted the primary endpoints from the included literature. Random-effect model or fixed-effect model were utilized to generate and compute relative risk and mean difference, as appropriate. Publication bias was quantified and assessed using the funnel plot with Egger's test. Results This study included 13 literatures with a total of 3047 pregnant women with gestational weeks more than 10 weeks. The incidence of vaginal delivery was significantly higher in the intervention group than that in the control group (28.7% vs 23.3%, P < 0.001). The differences of duration of the first stage and second stage of labor between the interventional group and control group were both statistically insignificant (mean difference: 27.92, 95% CI: − 70.60, 14.7, P = 0.20; mean difference: 0.63, 95% CI: − 4.47, 5.74, P = 0.81). In addition, there were no significant differences with regard to gestational age at delivery (mean difference = −0.23, 95% CI: − 1.29, 0.83, P = 0.67), Apgar score (mean difference = 0.06, 95% CI: − 0.13, 0.26, P = 0.53), and birth weight (mean difference = −23.78, 95% CI: − 60.66, 13.11, P = 0.21) between the 2 groups. Women in the intervention group were more likely to experience vaginal delivery than the control group (RR = 1.27, 95% CI: 1.04, 1.55, P = 0.01). Conclusions Physical exercise during pregnancy could improve the incidence of natural labor.