Induction of labor in twin pregnancies: systematic review and meta-analysis
Sara Sorrenti1, Can Ozan Ulusoy2, Doaa Mohammed3
1Department of Maternal and Child Health and Urological Sciences, Sapienza University of Rome, Italy; Fetal Medicine Unit, Liverpool Women's Hospital, Liverpool, United Kingdom.
Objectives:
To evaluate the maternal and perinatal outcomes of twin pregnancies undergoing induction of labor (IOL).
Methods:
MEDLINE, EMBASE, CINAHL, Cochrane database and the Scopus databases were searched electronically up to May 2024 (PROSPERO CRD42024520118). All studies regarding maternal or neonatal outcomes of induction of labor in twins were included. Language was restricted to English. Four different analyses were performed: IOL was compared to spontaneous onset of labor, elective Caesarean delivery, IOL in singletons and at last different methods of IOL in twin pregnancies were compared. Maternal and neonatal outcomes were analysed. Random-effect head-to-head meta-analyses were used to analyze the data. Quality assessment was performed using ROBINS-I for non-randomised and RoB2 for randomised studies.
Results:
The comparison of IOL and spontaneous onset of labor (8 studies) showed that the incidence of unplanned Caesarean delivery (CD) was similar between the two groups (pooled OR 0.94, 95% CI 0.36-2.45; p = 0.89). Neonatal deaths were significantly lower in the IOL group compared to the spontaneous onset of labor group (pooled OR 0.22, 95% CI 0.10-0.51, p = 0.0003 for Twin 1 and 0.29, 95% CI 0.14-0.60, p = 0.008 for Twin 2). On the other hand, the comparison of IOL with planned CD (7 studies) demonstrated that the rate of composite adverse maternal outcome was significantly higher in women undergoing IOL compared to women who had planned CD (pooled OR 2.48, 95% CI 1.16-5.31; p = 0.02). However, no significant differences were observed among these two groups regarding neonatal complications (p > 0.05). The comparison of IOL outcomes in twin and singleton pregnancies (3 studies) showed no significant difference in the rate of unplanned CD (pooled OR 1.42, 95% CI 0.89-2.28; p = 0.14). At last, the comparison of different methods of IOL (5 studies) showed similar rates of Caesarean deliveries were observed in the different groups (pooled OR 1.01, 95% CI 0.56-1.85; p = 0.97). The risk of bias assessment showed overall low risk of bias for observational studies (ROBINS-I) and some concerns for the randomized studies (RoB2). According to the GRADE system, the quality of evidence was overall low to moderate.
Conclusions:
IOL may be considered a safe option in twin pregnancies as it is significantly associated with a lower risk of neonatal mortality and does not seem to increase the risk of unplanned CD when compared to the spontaneous onset of labor. However, IOL is associated with higher maternal morbidity when compared to elective Caesarean delivery. Larger studies are needed to further evaluate predictive factors of unplanned CD after IOL among these pregnancies.
