3.3.6 Incidence of preterm premature rupture of the membranes
Four studies were included for this analysis[12,13,15,16]. Given the significant heterogeneity across studies (P =0.01, I 2=73%), sensitivity analysis was performed to detect the source of heterogeneity. After excluding the most influential study by Fichera 2019 et al. , heterogeneity was no longer significant (P =0.37, I 2=1%), and a fixed-effect model was used for analysis. A significant lower incidence of PPROM was found in the cervical pessary group compared with that in the cervical cerclage group (RR =0.48, 95%CI : 0.35 to 0.67,P <0.00001). Sensitivity analyses did not significantly alter the results, suggesting that the results were stable (Figure 1-7).