4. Discussion
In this cohort study of women pregnant with twins, delivering a gestation age between 32 weeks 0 days and 38 weeks 6 days with the first twin in cephalic position, we found that neonatal outcomes between VB and CB were similar in term pregnancies whereas adverse neonatal outcomes were increased in the VB group in preterm second twin.
In a large-scale retrospective study, it was concluded that vaginal twin delivery appears safe if experienced staff monitor birth weight differences, birth interval, and blood values (15). In a comprehensive randomized study, no difference was observed between fetal and neonatal mortality rates and severe neonatal morbidity when comparing VB and CB in twin pregnancies between 32 and 38 weeks with the first twin in vertex presentation (4,14). A recently published prospective cohort study found that an increased risk on adverse neonatal outcomes was seen after planned CB compared to planned VB at a gestation age of 32 weeks 0 days to 36 weeks 6 days. In addition, no difference was found between the groups after 37 weeks (10). In another study, no difference was observed between the groups for term pregnancies, but preterm pregnancies were found to have significantly higher rates in the VB group (13). Another study found that perinatal mortality did not differ statistically significant between planned CB and planned VB in preterm twins but at term, a planned CB may result in less asphyxia and trauma-related outcomes (16). In our study, there was no difference in ENMR in the term pregnancy groups when comparing CB and VB, but this rate was significantly higher in the VB group in the preterm group.
As for neonatal outcomes, it seems that second twins displayed higher morbidity rates (9,17,18). In our study, all the first twins were delivered via VB in group 1, and 17.97% of the second twins were delivered via emergency intrapartum cesarean sections, which is similar to the VB group in the study performed by Goossens et al. (intrapartum CB rate of 19.7%) (16). Early neonatal deaths occurred in two of the second twins who had to be delivered via emergency intrapartum cesarean sections in the VB group, and the ENMR in this group was 2.24%, while this rate was 0.97% in the CB group for the second twins. However, in the subgroup analysis, early neonatal deaths in both groups were observed in the preterm pregnant group. Unlike in our study, Barrett et al. (11), who reported no difference in terms of early neonatal mortality after CB and VB in the preterm pregnant group.
Zafarmand et al. (19) identified gestational age at birth as a strong prognostic factor for the outcomes of neonates, depending on the planned mode of birth. They also stated that from 32 to 37 weeks, a planned VB seems favourable, while from around 37 weeks on, a CB might be safer. In our study, early neonatal mortality was observed in both the VB and CB groups at 32–37 weeks but not over 37 weeks.
When VB is attempted, the capacity for immediate CB is important in the event that complications necessitating urgent birth arise (e.g., prolapsed umbilical cord, non-reassuring fetal heart rate, no descent of the fetal presentation, failed breech extraction, or failed internal podalic or external cephalic version, cervical retraction, prolapse of an arm and placental abruption). Our study showed that main indication for cesarean for the second twin was the failure of intrauterine manoeuvres. In the main, emergency situations presenting more than one obstetrical complication were responsible for this uncommon practice (20). Studies have shown that in twin pregnancies, the incidence of adverse perinatal outcomes in twin pregnancies delivered with emergency intrapartum cesarean sections is higher than in planned CBs (21-23). Our study also showed high ENMR results in twins with emergency intrapartum cesarean sections in the VB group, which is similar to the results of literature.
Grossman et al. (24) found that maternal morbidity increased in the VB group compared to the planned CB group. Also, in this study, the highest rate of adverse outcomes was seen in twins who underwent CB after failed induction of labor (24). Also, Mei-Dan et al. (14) found that planned VB group had more antepartum hemorrhage (1.9% vs 0.6%) and maternal complication (2.4% vs 0.1%) compared with the planned CB group. Conversely, in another study was found that in twin pregnancies with planned VB, CBs for the second twin and for both twins are associated with higher risks of severe acute maternal morbidity than VB (3). The multicentre retrospective study of Wenckus et al. (8) comparing maternal and neonatal outcomes in twins undergoing a trial of labor versus prelabor caesarean, there was an increased risk for postpartum haemorrhage and blood transfusion for the trial of labour. In our study, we did not find any difference in terms of maternal morbidity/mortality according to the mode of birth.
Although the literature is controversial, inter-twin weight discordance >20% was found to be a risk factor for increased perinatal morbidity of second twins (25). However, Peaceman et al. (26) emphasized that the route of birth does not influence neonatal outcomes when assessing weight discordance above 20%. In our study, there were no differences in terms of weight discordance in twin birth according to the mode of birth.
As a limitation, the small number of study population and retrospective collection of the data may be considered as weakness of our study. The strength of our study is that it evaluates a controversial issue.