3.3.9 Cesarean section rate
Four studies were included for this analysis[10,11,16,17]. Due to significant heterogeneity across studies (P =0.33, I 2=66%), a random-effect model was used. Meta-analysis did not find any significant difference in cesarean section rate between the two groups (RR =1.02, 95%CI : 0.72 to 1.44, P =0.93) (Figure 1-10).
3.4 Publication bias
A funnel plot was used to detect potential publication bias for the outcome of premature birth rate before 34 weeks. As pictorially shown in Figure 2, the funnel plot was symmetric on visual inspection, indicating a low risk of publication bias.
4. Discussion
Preterm birth is one of the most important determinants of perinatal and infant mortality and morbidity worldwide, which is among the most complex challenges to modern obstetrics. Spontaneous preterm birth is estimated to be responsible for roughly 28% of the world’s 4 million annual neonatal deaths[18]. Decreasing the incidence of spontaneous preterm birth is critical to reduce perinatal and neonatal mortality and morbidity. The onset and delivery of preterm labor is similar to that of full term, but the exact mechanism remains unknown. Short cervix syndrome, also known as cervical shortening, is considered as a strong predictor of preterm birth[19]. For pregnant women with a prior history of preterm birth or shortened cervix, secondary prevention based on proper management of a short cervix has been shown to reduce the risk of preterm birth[19].
However, the published studies on the use of cervical pessary versus cervical cerclage to prevent short cervix-related preterm birth are limited by their small sample size, and these studies reported conflicting findings. There is a lack of meta-analysis comparing cervical pessary with cervical cerclage in the management of preterm birth associated with a short cervix. Therefore, this study systematically reviewed the published RCTs and non-RCTs on the efficacy and safety of cervical pessary versus cervical cerclage in preventing spontaneous preterm birth, in order to offer more effective options for pregnant women with a short cervix who are at a risk for preterm birth. Our results indicated that compared with cervical cerclage, use of a cervical pessary significantly decreased the incidence of PPROM and premature birth rate before 34 weeks. But other endpoints including premature birth rates before 28, 32 or 37 weeks, cesarean section rate as well as neonatal outcomes were similar between the two groups.
We acknowledge that this study has several limitations: 1) Only 2 non-blinded RCTs were included, which are subject to potential selection bias and measurement bias. 2) One study was conducted in twin pregnancies while others were performed in singleton pregnancies. 3) Seven of the included studies used Arabin cervical pessary, while other studies did not specify the type of cervical pessary used; four studies used McDonald cervical cerclage, two used McDonald or Shirodkar cerclage, one used Lyubimova technique, one used Lyubimova or Mammadaliyeva technique, while other studies did not mention the specific technique used for cervical cerclage. The use of variable pessary types or cerclage techniques may influence outcome data. 4) Few studies reported all of these outcome measures and the quality of included studies varied, which may affect the reliability of our findings. 5) Few studies reported the potential complications (e.g. vaginal drainage, infection, bleeding, intrauterine infection, neonatal complications) or economic parameters (e.g. length of hospital stay, cost) associated with cervical cerclage and cervical pessary. Only one study reported the endpoint of mean gestational weeks extended. And thus, it is not possible to analyze these outcomes in this study.
In conclusion, compared with transvaginal cervical cerclage, use of a cervical pessary may decrease the risks of PPROM and premature birth before 34 weeks. Given its advantages of easy-to-use and minimal damage, cervical pessary may become a useful preventive intervention that deserves widespread clinical application. However, due to the limitations in the available evidence, more high-quality, large-scale, multicenter studies are needed to further clarify its role in preterm birth prevention.