The principal findings of the present study were: In women with dichorionic-diamniotic twin pregnancies delivered in the late preterm period, (1) ACT was not significantly associated with a reduction in adverse neonatal complications regardless of the timing of exposure during pregnancy and was significantly associated with an increased rate of neonatal hypoglycemia, particularly in women treated before 34 0/7 WG; (2) Rates of admission to NICU were significantly higher in the corticosteroid-exposed groups regardless of the timing of exposure during pregnancy; (3) Neonatal birthweight was significantly lower in the ACT < 34 0/7 WG group; (4) GAB, birthweight, and early preterm ACT were independent protective factors predicting CRO, CNO, and NICU admission.
In recent decades, various factors such as financial security, career priority, and lack of emotional and psychological readiness have discouraged women from becoming mothers at a young age. This trend has led to more and more women trying to become pregnant at an older age, when fertility begins to decline. Therefore, the number of pregnancies conceived using assisted reproductive methods is increasing, and as an inevitable consequence, the number of multiple pregnancies is also gradually increasing worldwide [21–24].
Multiple pregnancies are associated with an increased risk of pregnancy complications, particularly spontaneous abortions, preterm birth, preeclampsia, and maternal hemorrhage [21]. On the other hand, preterm birth is the leading cause of infant death, and the degree of prematurity is the most important determinant of morbidity and mortality [25]. Therefore, strategies to predict and prevent preterm birth and treatment approaches to reduce neonatal morbidity and mortality in populations at high risk of preterm birth are essential. In this context, fibronectin and, more commonly, sonographic measurement of cervical length have been used to predict preterm birth in singleton pregnancies [26]. In a recent study, the combination of sonographic measurement of cervical length and determination of granulocyte elastase in cervical secretions proved to be a useful tool for predicting preterm birth in asymptomatic twin pregnancies [27].
Administration of corticosteroids to women in whom preterm birth is inevitable is the most important antenatal intervention to reduce adverse neonatal outcomes [4, 28]. ACT is recommended between 24 0/7 and 33 6/7 WG and may even be considered between 23 0/7 and 24 0/7 WG in women with an inevitable preterm birth within seven days, including pregnancies with rupture of membranes [4, 7, 29]. Recent data suggest that ACT may be beneficial in women at risk of late preterm delivery [28]. The risk of neonatal sepsis, chorioamnionitis, or endometritis was not increased, but neonatal hypoglycemia occurred more frequently when ACT was administered in the late preterm period [30].
On the other hand, there are vulnerable pregnancy populations, including twins and multiple pregnancies, where the effects of ACT have not yet been adequately studied. As there is a need for studies that address the gaps and practice for the use of late preterm ACT in women with dichorionic-diamniotic twins who are at high risk for inevitable late preterm birth, we conducted this article. Furthermore, to our knowledge, this article is the first study to examine the effects of ACT in such pregnancies, comparing whether they were treated before or after 34 0/7 WG. A study conducted by Gyamfi-Bannerman et al. found that late preterm corticosteroid therapy was associated with improved respiratory morbidity in late preterm infants [28]. In another study by Martinka et al, which was designed as a secondary analysis of an earlier randomized controlled trial in twin pregnancies (Twin Delivery Study) and in which the analysis was restricted to women who had delivered in the late preterm period, the rates of improvement in composite respiratory outcomes were comparable to the corresponding rates in the study by Gyamfi-Bannerman et al. The authors concluded that the benefit-risk ratio of ACT with respect to neonatal respiratory morbidity in twin pregnancies at risk of late preterm delivery is expected to be similar to that in singletons [31, 32]. However, this conclusion is limited because it is a theoretically based indirect assumption and the pharmacodynamics of corticosteroids are different in singleton and twin pregnancies. Twins develop in a different environment than singletons and face many difficulties during the intrauterine period, such as competition for maternal nutrients, limited space and smaller placentas or even smaller placental proportions in the case of monochorionicity.
In a study by Ben-David et al. similar results were obtained as in our study regarding neonatal outcomes [33]. However, they compared only the women exposed to ACT during late preterm period with the control group, whereas the comparison in our study was made by dividing the treatment groups according to whether ACT was administered before or ≥ 34 0/7 WG. In addition, a more homogeneous distribution was achieved in our study by excluding diabetic pregnancies and twins other than dichorionic-diamniotic. Corticosteroid treatment is a double-edged sword because, although it reduces the risk of respiratory neonatal morbidity and mortality, steroid-induced maternal hyperglycemia can also lead to other neonatal outcomes, particularly neonatal hypoglycemia, especially in preterm newborns. Since maternal glucose supply is interrupted immediately after birth, the newborn must initiate endogenous glucose production to maintain euglycemia. This adaptation process begins with a decrease in neonatal insulin and an increase in glucagon shortly after birth [34]. This mechanism may be disrupted in hyperinsulinemic preterm newborns exposed to ACT, and persistent neonatal hypoglycemia can lead to serious consequences [35, 36]. Fortunately, maternal hyperglycemia induced by corticosteroids lasts only a few days [37]. On the other hand, a recent study by Carpenter et al. found an increase in granulocytes and a decrease in lymphocytes in the umbilical cord blood of fetuses whose mothers had received betamethasone treatment during late preterm period [38]. These alterations in the white blood cells can have negative consequences for the newborn, which tries to adapt to the external environment and fight possible infectious agents after the protective maternal immunity has been extinguished at birth.
In our study, neonatal hypoglycemia rates were significantly higher in the ACT < 34 0/7 WG group and were observed more frequently in the ACT ≥ 34 0/7 WG group than in the unexposed group. On the other hand, it should be considered that this result could be closely related to the significantly higher GAB in the control group, in which the protective mechanisms against hypoglycemia function better than in those born at earlier weeks. Mean birthweight was also significantly lower in the ACT < 34 0/7 WG group than in the control group, but there was no significant difference between the ACT ≥ 34 0/7 WG group and the control group. In addition, the rates of NICU admission and need for mechanical ventilation were higher in the ACT < 34 0/7 WG group. Similar to hypoglycemia, all these results could be closely related to the significantly lower GAB in this group. Although the APGAR-5 values in the ACT ≥ 34 0/7 WG group were statistically significantly higher than in the other groups, this result loses significance as the median values are very close to each other and the minimum APGAR-5 score was already seven.
In the light of the results of our study, it would be appropriate not to perform ACT in twins diagnosed with preterm labor unless birth is predicted before 34 0/7 WG. Given the known side effects and as yet unknown effects of corticosteroids, it is therefore very important to accurately identify women in whom the risk of preterm birth before 34 0/7 WG is unavoidable [39, 40]. The main limitation of our study is its retrospective nature. In addition, intrauterine physiology and pharmacodynamics differ in multiple pregnancies, and the corticosteroid doses in our study, as in many studies, were the same standard doses used in singleton pregnancies. The major strength is that the study was conducted in a large tertiary referral hospital where the same algorithms were used for the diagnosis, management and follow-up of women with twin pregnancies.