Small for Gestational Age in Late Preterm Deliveries - The Impact of Antenatal Corticosteroids on Perinatal Outcome

Purpose The aim of this study is to evaluate the impact of antenatal corticosteroids (ACS) on late preterm small for gestational age neonates. Methods A retrospective cohort study of all women, carrying a singleton gestation, who had late preterm delivery (34 + 0–36 + 6, gestational weeks) of small for gestational age neonates, in a single, tertiary, university-aliated medical center (July 2012- December 2017). Exclusion criteria included: birth weight above the 10th percentile, termination of pregnancy and intrauterine fetal death. Outcomes were compared between those who were treated with ACS prior to delivery and those who did not receive ACS. The primary outcome was neonatal composite outcome which included: neonatal intensive care unit admission, respiratory distress syndrome, mechanical ventilation and transient tachypnea of the newborn.


Introduction
Antenatal corticosteroid (ACS) therapy for preterm infants has been shown to improve neonatal outcome.
Back in 1972, Liggins et al. demonstrated that ACS therapy reduces respiratory distress syndrome (RDS) in preterm deliveries [1]. The latest Cochrane review which included the results of 30 trials, concluded that a single course of a corticosteroids, given to the parturient in preterm labor, reduces the rates and severity of serious adverse outcomes related to prematurity, including: RDS, intraventricular hemorrhage (IVH), necrotizing enterocolitis (NEC), need for respiratory support and neonatal intensive care unit (NICU) admission [2]. In addition, ACS treatment was found to signi cantly reduce the risk of perinatal and neonatal death. The impact of ACS on late preterm neonates was assessed in a recent randomized trial and found that antenatal administration of betamethasone to women at risk for late preterm delivery (34 weeks 0 days to 36 weeks 6 days of gestation) decreased the need for substantial respiratory support during the rst 72 hours after birth [3].
Notwithstanding the above, ACS are associated with several adverse effects especially when repeated courses are given. Murphy et al. demonstrated that multiple courses of ACS, every 14 days, do not improve preterm-birth outcomes and are associated with a decreased birth weight, length and head circumference at birth [4]. In a another study that evaluated long term outcomes after repeated doses of ACS, where children were followed up to 2-3 years of age, found higher rate of cerebral palsy among children who had been exposed to repeated doses of corticosteroids [5].
Although ACS are widely used, there are several unanswered questions regarding this treatment, one of which is the impact of this treatment on growth restricted fetuses [6]. It is well established that SGA contributes to neonatal morbidity in preterm and term neonates as it contributes to higher risk for premature complications including RDS, IVH and NEC, and also for long term adverse outcomes such as cerebral palsy, major psychiatric sequelae in later years and adult cardiovascular diseases [7]. In one study, neonatal outcome was compared between growth restricted fetuses that received ACS and those who did not receive ACS, they concluded that administration of corticosteroid to growth restricted preterm fetuses does not appear to be bene cial with respect to short term neonatal outcome [8]. However, this study included only early preterm (up to 34 weeks of gestation) growth restricted fetuses. To the best of our knowledge, little is known on the impact of ACS on SGA neonates born in late preterm. Hence, our research aimed to assess whether exposure to ACS during pregnancy improves adverse neonatal outcome in late preterm SGA.

Study population
A retrospective cohort study of all women, carrying a singleton gestation, who had a preterm delivery in a single, tertiary, university a liated medical center between July 2012 and December 2017.
We only included late preterm deliveries (34+0-36+6 gestational weeks) who had small for gestational age newborns de ned as birth weight (BW) below the 10th percentile, according to the Israeli national birthweight curves [9].
Exclusion criteria included: termination of pregnancy (TOP), intrauterine fetal death (IUFD) and birth weight at or above the 10 th percentile.

Ethical approval
The study was approved by the institutional review board of Rabin Medical Center (RMC-19-0557).
Informed consent was waived due to the retrospective design of the study.

Data collection
Data were retrieved from the comprehensive computerized perinatal database of our center. Data from the neonatal unit and the neonatal intensive care unit (NICU) were integrated into the delivery room database using the unique admission number assigned to each woman and her offspring. Collected data included demographic and obstetric parameters, labor and short-term maternal and neonatal outcome (up to discharge).

Outcome measures
The study population comprised two groups, those who were treated with ACS prior to delivery and those who did not receive ACS prior to delivery. Maternal and neonatal outcomes were compared between groups.
The primary outcome was composite neonatal outcome which included: NICU admission, RDS, mechanical ventilation and transient tachypnea of the newborn (TTN).
Secondary outcome was adverse neonatal outcome, adverse maternal and labor outcome. Neonatal adverse outcome included: birth weight, gender, umbilical arterial PH, RDS, TTN, IVH, mechanical ventilation, NEC, retinopathy, sepsis, antibiotic treatment, NICU, jaundice requiring phototherapy, neonate major anomaly, blood products transfusion and hypoglycemia. Adverse maternal and labor outcome included: postpartum hemorrhage (PPH), blood product transfusion, mode of delivery and onset of labor (elective, spontaneous or induction).
ACS was delivered in cases with suspected preterm delivery. By departmental protocol, the treatment course includes two 12-mg doses of betamethasone given intramuscularly 24 hours apart. In few cases, where pregnancy was continued and there was an imminent threat of preterm delivery, additional rescue course was given in the same manner.
Delivery methods of induction were prostaglandin E2 -PGE2, extra-amniotic balloon, and oxytocin infusion, which were chosen according to the physician's discretion and local institutional practice.

Statistical analysis
Continuous variables were compared using Mann-Whitney test. Correlations between continuous variables were evaluated using the Spearman correlation coe cient. The Chi squared test was employed to compare categorical variables. Logistic regression analysis was used to determine which factors were signi cantly and independently associated with ACS treatment.
Odds ratios (OR) with 95% con dence interval (CI) were reported. All statistical tests were 2-tailed, and p < 0.05 was considered as statistically signi cant. All calculations were performed using IBM SPSS (Ver. 26.0).

Results
Overall, 228 patients met inclusion criteria. Of them, 102 (44.7%) received ACS and 126 did not (55.3%). Among the group that received ACS, 27 (26.47%) received a second (rescue) course of ACS. Maternal characteristics are summarized in Table 1.
There were no differences between groups regarding maternal age, body mass index, gravidity and parity. There were signi cantly higher rates of gestational diabetes and lower rates of preeclampsia among women who received ACS (9.8% vs. 2.38%, p = 0.01 and 9.8% vs 19.9%, p=0.03, respectively). Regarding obstetric outcomes, the group that received ACS delivered earlier (35.5 vs. 36.1 gestational weeks, P = 0.00).
Aa for onset of labor, in the non-ACS group, there was a higher prevalence of spontaneous delivery versus elective delivery and induction of labor (Table 2).
A sub-analysis was performed for the group that received ACS. There was a higher rate of NICU admission among the group that received a rescue course of ACS (70.37% vs 47.14%, p=0.04). In addition, composite neonatal outcome was signi cantly higher among women that received a rescue course of ACS (70.37% vs 47.14%, OR 2.66, CI 1.03-6.88, p=0.04). Following regression analysis this association remained signi cant (OR 2.65, CI 1.01-6.91, p=0.04).

Discussion
In this study we found that administration of ACS during pregnancy does not appear to be bene cial on short term outcome and can even result in higher rates of adverse outcome in SGA neonates born in the late preterm. Our study showed that the rate of NICU admission was signi cantly higher among the ACS group along with higher rates of composite neonatal outcome. This association remained signi cant among fetuses that received a rescue course of ACS in sub-analysis for the ACS group only.
ACS therapy for decreasing neonatal morbidity and mortality is recommended by guidelines around the world and is widely used. Its bene ts among intrauterine growth restriction (IUGR) fetuses remains largely unknown, and controversy exists on the bene t of ACS for IUGR/SGA fetuses to improve preterm birth outcome. There is insu cient evidence to withhold routine ACS therapy in cases of suspected IUGR with a high risk of preterm birth.
The bene t of ACS in speci c obstetric population such as SGA neonates, is yet to be determined. A previous study by Gyam -Bannerman et al., investigated the effects of ACS for women at risk for late preterm delivery. In this randomized trial, women with a singleton pregnancy who were at high risk for delivery during the late preterm period were recruited. In this study ACS signi cantly reduced rates of neonatal respiratory complications. It should be noted though that the frequency of IUGR among the ACS group was 3.2% and among the control group was 3.4%, and the impact on ACS on this sub group was not analyzed [3]. Another study by Haviv et al., investigated the role of ACS on late preterm in special populations. They concluded that there is insu cient evidence regarding the bene ts or harms of ACS therapy in pregnancies with IUGR, especially in the late preterm period. They recommended an individualized approach when administering corticosteroids at later gestations in speci c obstetric populations such as IUGR [10].
Several studies reported no effect of ACS on neonatal morbidity or mortality among IUGR fetuses in the early preterm (up to 34 weeks of gestational age) [8,11,12,13,14]. Van Stralen et al. demonstrated that administration of ACS to IUGR fetuses does not appear to be bene cial with respect to short term neonatal outcome in preterm deliveries [8]. Another recent study showed the same results where ACS did not improve neonatal morbidities, in SGA neonates delivered between 29 and 34 gestational weeks. Rather, ACS seemed to increase the risk of RDS. They concluded that ACS therapy for women who are at risk for preterm delivery with IUGR fetus, need to be further evaluated, espe cially after 32 weeks of gestation [13]. A recent meta-analysis, examined 16 observational cohort and case-control studies published from 1995 to 2018, they concluded that ACS reduces neonatal mortality in SGA infants delivered preterm, with no apparent effect on neonatal morbidity (RDS, NEC, IVH and periventricular leukomalacia, bronchopulmonary dysplasia or chronic lung disease of prematurity, or neonatal sepsis). The study concludes that future studies are required on the effect of ACS administration to SGA infants in the late preterm period, because data on this issue is limited [14] One assumption for our results is that poor intrauterine growth, by itself, actually enhances lung maturation. This assumption has been demonstrated in several studies. The physiological adaptations that growth restricted fetus experience in response to nutrient and oxygen restriction alter the ability to regulate endogenous glucocorticoid availability. As a result, these fetuses may be exposed to higher ACS concentrations, which may result in an exacerbation of the potentially negative side effects of antenatal glucocorticoid treatment, especially in cardiovascular development. Possibly without the full capacity to bene t from the lung maturational effects [15]. Conversely, a previously published study demonstrated that IUGR fetuses accelerate lung maturation is not supported in comparisons of SGA and appropriate for gestational age (AGA) infants of the same gestational age, sex and race [16].
Secondly, it has been suggested that elimination of ACS via the placenta or the blood-brain barrier is impaired with IUGR, and hence, the fetus is exposed to excessive corticosteroids in the lung, brain, and heart tissues [15].
Nevertheless, some studies showed lower risk of adverse outcomes [16,17,18,19,20]. Bernstein et al. demonstrated the association of IUGR fetuses with increased morbidity and mortality. Furthermore, it showed that the bene ts of ACS therapy were similar among infants with IUGR and normally grown infants for neonates from 25 to 30 weeks of gestation [19]. A population-based study on singleton infants of 24-31 weeks of gestation, concluded that ACS therapy was associated with signi cantly reduced mortality and major neonatal morbidities among preterm SGA neonates which was generally similar to the effect in the AGA preterm infants [17]. A previous review (2018) concluded that based on the current clinical evidence, it is reasonable to give a single course of glucocorticoids to pregnant women with IUGR fetuses who are at risk of preterm birth, however there is insu cient evidence to conclude whether repeated or rescue ACS administration is bene cial for IUGR infants [18].
We also found that birth weights were signi cantly higher among non-ACS group. Our results are aligned with previous studies which have shown that ACS is associated with reduction in birth size for infants born preterm, near term, or at term [21]. These studies even showed reduction in head circumference among preterm newborns [21,22]. However, it is unclear whether this difference in birth weight is a result of the ACS treatment or was it the reason for administering the ACS.
The novelty of the present study is that we examined ACS therapy for SGA fetuses eventually born at late preterm. To the best of our knowledge, all existing studies examined ACS therapy for SGA fetuses in the early preterm. Moreover, most studies refer to IUGR fetuses (de ned as an estimated fetal weight <10th percentile) and not SGA fetuses.
Nevertheless, it is not free of limitations. The main limitation of this current study is its retrospective design, which could lead to an unknown selection bias such as the reason for administering or withholding ACS. In addition, the timing of administering ACS was missing. Another limitation is that we only studied short-term neonatal outcomes in this speci c population, and the long-term impact of ACS is yet to be determined.
In conclusion, our study demonstrated that ACS did not decrease neo natal morbidity, in SGA neonates at the late preterm. It might be even associated with adverse neonatal outcome. This should be further evaluated in large prospective studies.