Effect of delayed cord clamping on maternal and neonatal outcome in twin pregnancies: a retrospective cohort study

The objective of this study was to compare the maternal and neonatal outcomes following delayed cord clamping (DCC) versus immediate cord clamping (ICC) in twin pregnancies. This was a retrospective cohort study of 705 twin pregnancies who delivered at ≥ 24 weeks of gestation. Maternal and neonatal hemoglobin levels, blood transfusion, and neonatal outcomes were compared between DCC (n = 225) and ICC (n = 480) groups. Mean maternal predelivery and postpartum hemoglobin levels and the rate of postpartum hemoglobin drop ≥ 20% or maternal blood transfusion were comparable between the two groups. The DCC group had a significantly higher mean neonatal hemoglobin level (DCC vs. ICC: 17.4 ± 3.5 vs. 16.6 ± 2.7 g/dl, P = 0.010) but significantly lower rates of neonatal blood transfusion (DCC vs. ICC: 3.3% vs. 8.8%, P < 0.001) and respiratory distress syndrome (DCC vs. ICC: 6.7% vs. 15.2%, P < 0.001) than the ICC group. In conclusion, DCC compared with ICC in twin pregnancy was not associated with an increase of maternal postpartum bleeding complications, but it was associated with higher neonatal hemoglobin level and lower risks of neonatal blood transfusion and respiratory distress syndrome.


Ethical approval
This study was approved by the Institutional Review Board for Clinical Research at Samsung Medical Center at December 22, 2022 (IRB No. 2022-12-101).

Results
During the study period, 754 women with twin pregnancies delivered at ≥ 24 weeks of gestation in our institute.After excluding 49 women based on our exclusion criteria, a total of 705 twin pregnant women were included in the final analysis (Fig. 1).DCC was performed in 225 (31.9%) of 705 women.
Maternal characteristics including age, BMI, ART pregnancy, and chorionicity were comparable between the two groups (Table 1).However, the proportion of multiparous women was significantly higher in the ICC group than in the DCC group.Rates of pregnancy complications including preterm labor, IIOC, preeclampsia, placenta previa, placenta abruption, and gestational diabetes were similar in the two groups (Table 2).However, the rates of antenatal corticosteroids treatment and PPROM were significantly higher in the ICC group than in the DCC group.Gestational age at delivery was significantly higher in the DCC group than in the ICC group.Rates of PTD and cesarean delivery were significantly higher in the ICC group than in the DCC group.Mean  www.nature.com/scientificreports/maternal predelivery hemoglobin levels were similar between the two groups (Table 4).Mean maternal postpartum hemoglobin levels on the first day (DCC vs. ICC: 10.3 ± 1.6 vs. 10.5 ± 1.6 g/dl, P = 0.111) and the third day (DCC vs. ICC: 9.9 ± 1.5 vs. 10.1 ± 1.5 g/dl, P = 0.346) were also comparable between the two groups.Rates of postpartum hemoglobin drop ≥ 20% and maternal blood transfusion were not significantly different between the two groups.In subgroup analyses, mean maternal predelivery and postpartum hemoglobin levels (day 1 and day 3) were also similar between DCC and ICC groups (Table 5).However, the rate of postpartum hemoglobin drop ≥ 20% was significantly higher in the DCC group than in the ICC group for those with a preterm delivery.Birth weight was significantly higher in the DCC group.However, proportions of SGA, appropriate for gestational age (AGA), and LGA were not significantly different between the two groups (Table 3).There was no difference in the proportion of those with a 5-min Apgar score below 7 between the two groups.Neonatal hemoglobin levels were available for 151 and 433 cases in DCC and ICC groups, respectively.Mean neonatal hemoglobin level was significantly higher in the DCC group than in the ICC group (DCC vs. ICC: 17.4 ± 3.5 vs. 16.6 ± 2.7 mg/dl, P = 0.010) (Table 4).However, in subgroup analyses, mean neonatal hemoglobin level was significantly higher in the DCC group than in the ICC group only in the cesarean section group (Table 5).Neonatal blood transfusion data were available for all neonates.Neonatal blood transfusion rate was significantly lower in the DCC group than in the ICC group (3.3% vs. 8.8%, P < 0.001) (Table 4).In subgroup analyses, neonatal blood transfusion rate was significantly lower in the DCC group than in the ICC group in the preterm delivery group, cesarean section group, and dichorionic twin group (Table 5).In the subgroup analyses to compare the neonatal hemoglobin level and blood transfusion of according to the twin birth order (DCC 1st twin vs. ICC 1st twin, and DCC 2nd twin vs. ICC 2nd twin), overall results were similar to those of all twins (Table 6).www.nature.com/scientificreports/There was no significant difference in the rate of IVH (≥ grade 3), PVL, hypoglycemia, hyperbilirubinemia, sepsis, or death between the two groups (Table 4).The rate of RDS was significantly lower in the DCC group than in the ICC group (DCC vs. ICC: 6.7% vs. 15.2%,P < 0.001, adjusted odds ratio: 0.517, 95% confidence interval: 0.276, 0.970) after controlling for parity, PPROM, chorionicity, antenatal corticosteroids treatment, gestational age at delivery, cesarean section, twin birth order, sex, birth weight, and SGA (Table 7).Rates of NICU admission, mechanical ventilator therapy, BPD, NEC (≥ stage 2), and ROP (≥ stage 3) were lower in the DCC group than in the ICC group.However, these were not statistically significant in the multivariable analysis (Table 7).

Discussion
In this study, we evaluated maternal and neonatal outcomes following DCC versus immediate cord clamping (ICC) in twin pregnancies.Our study demonstrated that DCC in twin pregnancy did not result in increased maternal postpartum blood loss, although it was associated with significantly higher neonatal hemoglobin levels, lower neonatal blood transfusion rate, and reduced risk of RDS compared with ICC.
Although DCC is known to be more beneficial for neonates than traditional ICC, DCC might also have potential risks of maternal or neonatal adverse effects such as increased maternal blood loss, neonatal polycythemia, hyperbilirubinemia, and jaundice 1,11,17,20 .As for maternal bleeding complications, theoretically, DCC might potentially cause more maternal blood loss at the uterine incision site or episiotomy site.In particular, DCC might result in greater maternal blood loss in twin pregnancies because DCC takes even longer time from delivery to incision site repair than single pregnancies.
In previous studies of singleton pregnancies, DCC did not increase the risk of maternal bleeding complications such as increased estimated blood loss, higher postpartum hemorrhage rate, higher maternal blood transfusion rate, lower postpartum hemoglobin levels, or greater mean hemoglobin changes 1,2 .In a randomized controlled trial of multiple-birth infants born preterm at 28-36 weeks of gestation, rate of postpartum hemorrhage (defined as an estimated blood loss of > 500 ml for vaginal delivery or > 1000 mL for cesarean delivery) was significantly higher in the DCC group than the ICC group (6/24 [25%] in DCC vs. 1/23 [4.3%] in ICC; P = 0.04) 7 .However, the sample size of that study was too small (24 and 23 mothers in DCC and ICC groups, respectively), with triplet pregnancies (7/47) included in that study.In a retrospective cohort study including 449 multiple pregnancies, there were no significant differences in maternal bleeding complications (including postpartum hemorrhage, estimated blood loss, maternal blood transfusions, therapeutic hysterectomy) between DCC and ICC groups 13 .That study was similar to our study because the sample size was large, the rate of higherorder pregnancies was relatively low (2/154 in DCC vs. 9/295 in ICC), and both preterm and full-term infants www.nature.com/scientificreports/were included.In our study, postpartum hemoglobin levels tended to be slightly lower in the DCC group than in the ICC group, although such differences were not statistically different.In addition, the rate of postpartum hemoglobin drop ≥ 20% was slightly higher in the DCC group than in the ICC group.It was only statistically significant in the preterm delivery group.However, severe maternal postpartum hemorrhage leading to blood transfusion was not significantly different among all subgroups except the preterm delivery group.The reason for a higher rate of postpartum hemoglobin drop ≥ 20% in the preterm DCC group is not certain.It might be explained that most of the twin deliveries at preterm gestation are performed emergently and lower uterine segment is underdeveloped at preterm gestation which can result in more bleeding at the uterine incision site during cesarean section, especially when the uterine closure is delayed due to DCC.Another retrospective study of 82 twin pregnancies delivered at < 32 weeks of gestation showed that DCC was associated with higher estimated maternal blood loss in the cesarean section group, although maternal complications including maternal hemoglobin decrease, postpartum hemorrhage, blood transfusions, and hysterectomy were comparable between DCC and ICC groups 6 .Previous studies have well demonstrated that DCC could increase neonatal hemoglobin levels, improve iron status, and reduce risk of neonatal morbidities in singleton pregnancies 1,3,4,8 .However, the effect of DCC on neonatal hemoglobin level in twin pregnancies was controversial in previous studies.A randomized controlled trial showed that neonatal hemoglobin levels were similar in preterm twins or triplets infants who received DCC (n = 51) or ICC (n = 50) 7 .A retrospective study of dichorionic twin pregnancies at 23-32 weeks of gestation also reported that neonates who received DCC had no difference in neonatal hemoglobin level 14 .However, only 8 twin pregnancies (16 neonates) were included in the DCC group.In other studies, twins who received DCC had higher hemoglobin levels but lower rates of blood transfusion 6,12,15,16 .In our study, mean neonatal hemoglobin www.nature.com/scientificreports/level was higher while neonatal blood transfusion rate was significantly lower in the DCC group than in the ICC group in all study population.In subgroup analyses, neonatal hemoglobin levels in all subgroups tended to be higher in the DCC group than in the ICC group.However, neonatal hemoglobin level was significant only in the cesarean section group.This was probably because the sample size of our study, especially the numbers of subjects in each subgroup, was too small.In our institute, we do not perform routine complete blood cell count test for healthy babies who are not admitted to the NICU.Thus, neonatal hemoglobin level data were available in only 151 (33.6%) of 450 in the DCC group and 433 (45.2%) of 962 in the ICC group.This is one of the main limitations of our study.However, neonatal blood transfusion data were available in all neonates.Its rate was significantly lower in the DCC group than in the ICC group, especially in twins delivered at preterm, twins delivered by cesarean section, and dichorionic twins.www.nature.com/scientificreports/Other benefits of DCC include reduced risk neonatal morbidities such as respiratory distress syndrome, intraventricular hemorrhage, necrotizing enterocolitis, and death or major disability 1,[3][4][5]16,21 . Howevr, most of these studies were done in singleton pregnancies.There are only a few studies on twin pregnancies 6,7,12,[14][15][16] , and the results were controversial.A prospective cohort study of 202 twin pregnancies at > 32 weeks of gestation showed that twins who received DCC were at a lower risk of respiratory disorders and NICU admission 16 .Another retrospective cohort study including twin pregnancies at < 30 weeks of gestation showed a shorter NICU length of stay in DCC twins 15 .However, other studies of twin pregnancies found no differences in mechanical ventilator treatment, RDS, BPD, IVH, NEC, ROP, sepsis, death, or severe neurologic injury 6,7,12,14 .In our study, rates of NICU admission, mechanical ventilator treatment, RDS, BPD, NEC, and ROP were lower in the DCC group than in the ICC group, but this may be due to a lower rate of PTD in the DCC group.The rates of NICU admission, mechanical ventilator treatment, BPD, NEC, and ROP were not significantly different between the two groups after adjusting for confounding variables including gestational age at delivery.However, the rate of RDS was significantly lower in the DCC group in the multivariable analysis. The exact reson for reduced risk of RDS in babies who received DCC is unclear.Previous studies have suggested that DCC may improve hemodynamic stability and reduce complications related to poor oxygenation [22][23][24][25][26] .Chiruvolu et al. reported that the incidence of RDS and surfactant administration decreased when DCC is performed in preterm singleton pregnancies 4,5 .They explained that spontaneous respiration resulting in lung aeration starts during DCC, and thus the need for resuscitation decreases due to better cardiopulmonary transition to extrauterine life.Especially in premature infants, gas exchange is difficult due to high pulmonary vascular resistance 27,28 .In a study using the computational lumped parameter model of the placental and respiratory system, it was found that preterm infants who received ICC had lower blood volume, lower cardiac output, and lower blood pressure, whereas preterm infants who received DCC had higher oxygen saturation of the carotid and pulmonary arteries immediately after birth when DCC was performed 29 .However, the sample size of our study as well as other previous studies might be insufficient to conclude effects on neonatal outcomes other than neonatal hematologic effects in twin pregnancies.
The strength of this study was that it had a relatively large sample size of 705 twin pregnant women (1410 neonates) compared to previous twin studies.Another strength was that we recruited all twin pregnancies including both preterm and term gestation, both monochorionic and dichorionic twins, and both vaginal and cesarean delivery, while previous studies only included preterm twins or dichorionic twins.However, our sample size was still insufficient to have an adequate power because numbers of subjects in each subgroup were too low.In addition, neonatal hemoglobin level data were available in only less than half of all twin neonates.This study is further limited by the inherent nature of a retrospective study design.Maternal baseline characteristics and pregnancy outcomes were not comparable.Especially, more preterm twin pregnancies were included in the ICC group and cesarean section rate was higher in the ICC group.Although we performed subgroup analyses and multivariable analyses to control for bias, there might be other unknown potential confounding factors including individual surgeons' operative skill.There was no common protocol of umbilical cord clamping in our institute and DCC or ICC was performed at physicians' discretion.Individual surgeons' operative skills might be one of the unknown potential confounding factors.However, we were not able to adjust this because there were more than 30 doctors (including residents, fellows and professors) who delivered babies (vaginal or cesarean) during 8 years of study period.

Conclusion
Delayed umbilical cord clamping in twin pregnancy was not associated with maternal postpartum bleeding complications, although it was associated with increased neonatal hemoglobin level and decreased risk of RDS.However, more well-designed studies with larger sample size are needed to identify the benefit and risk of DCC in twin pregnancies.

Figure 1 .
Figure 1.Flow chart showing the selection of patients in this study.a Among 49 cases, 2 cases met the 2 exclusion criteria and 1 case met the 3 exclusion criteria at the same time.

Table 1 .
Maternal characteristics.Data are presented as number (percentage) or mean ± standard deviation.BMI body mass index, ART assisted reproductive technology.

Table 2 .
Pregnancy outcomes.Data are presented as number (percentage) or mean ± standard deviation.PPROM preterm premature rupture of membranes, IIOC incompetent internal os of cervix.a Analyzed only cases with data available on gestational diabetes.

Table 3 .
Neonatal characteristics.Data are presented as number (percentage) or mean ± standard deviation.SGA small for gestational age, AGA appropriate for gestational age, LGA large for gestational age.

Table 4 .
Primary and secondary outcomes.Data are presented as number (percentage) or mean ± standard deviation.Hb hemoglobin, NICU neonatal intensive care unit, RDS respiratory distress syndrome, BPD bronchopulmonary dysplasia, TTN transient tachypnea of the newborn, IVH intraventricular hemorrhage, PVL periventricular leukomalacia, NEC necrotizing enterocolitis, ROP retinopathy of prematurity.a Analyzed only in babies who had hemoglobin exam after birth (151 in the delayed cord clamping group and 433 in the immediate cord clamping group).

Table 7 .
Multivariable regression analysis of delayed cord clamping vs. immediate cord clamping.aOR adjusted odds ratio, CI confidence interval, NICU neonatal intensive care unit, RDS respiratory distress syndrome, BPD bronchopulmonary dysplasia, NEC necrotizing enterocolitis, ROP retinopathy of prematurity.a Adjusted for multiparity, preterm premature rupture of membranes, antenatal corticosteroids treatment, chorionicity, gestational age at delivery, cesarean section, twin birth order, sex, birth weight, and small-forgestational age.