Antenatal corticosteroid administration-to-delivery intervals and neonatal outcomes in late preterm labor: a historical cohort study

Abstract

Background

While antenatal corticosteroids (ACSs) are standard for early preterm birth, their benefit‒risk profile for the late preterm period is controversial, and the effect of the ACS-to-delivery interval on key neonatal morbidities is unclear. This study therefore examined the association between this interval and neonatal outcomes.

Methods

A historical cohort study enrolled 788 singleton pregnancies at 34^+ 0 to 36^+ 6 weeks of gestation. The non-ACS group comprised women who did not receive ACS. The ACS group was stratified into three subgroups according to the time between the first ACS dose and delivery: less than 2 days, between 2 and 7 days, and more than 7 days. The primary outcome was neonatal respiratory distress syndrome (NRDS), and neonatal hypoglycemia was a key secondary outcome. The associations between newborn outcomes and the intervals between ACS treatment and delivery were examined via multivariable logistic regression.

Results

Overall, ACS administration was not associated with reduced NRDS risk. However, this association significantly varied by the administration-to-delivery interval. Compared with the non-ACS group, a significantly lower NRDS risk was observed only in the 2–7 days subgroup [odds ratio (OR) = 0.468, 95% confidence interval (CI): 0.261–0.838, P = 0.011], but not in the < 2 days or > 7 days subgroups (both P > 0.05). In contrast, ACS use was associated with a significantly higher overall risk of neonatal hypoglycemia (OR = 2.594, 95% CI: 1.359–4.951, P = 0.004). This elevated risk was observed in both the < 2 days (OR = 2.922, 95% CI: 1.459–5.853, P = 0.002) and 2–7 days subgroups (OR = 2.641, 95% CI: 1.228–5.681, P = 0.013), but not when the interval exceeded 7 days. For other secondary outcomes, the need for mechanical ventilation was significantly reduced in the 2–7 days group (OR = 0.365, 95% CI: 0.152–0.873, P = 0.024).

Conclusions

In late preterm labor, ACS administration 2–7 days before delivery was associated with a lower risk of NRDS, whereas administration within 7 days was associated with a higher risk of neonatal hypoglycemia. These interval-dependent associations highlight the potential influence of timing on the benefit-risk profile of ACS and indicate the need for prospective validation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12884-026-08961-1.

Introduction

Preterm births between 32 and 37 weeks of pregnancy accounted for more than 85% of all preterm births, with a global prevalence of 9.9% in 2020 [1]. Within this group, late preterm births (defined as 34^{+ 0} to 36^{+ 6} weeks of gestation) account for a substantial share. Compared with full-term infants, these late preterm infants are far more likely to experience respiratory issues and neonatal hyperbilirubinemia [2].

To promote fetal lung development and improve outcomes for preterm infants [3], international clinical guidelines consistently recommend giving antenatal corticosteroids (ACSs) to pregnant women at risk of delivering prematurely before 34 weeks of gestation. Additionally, a landmark randomized controlled trial (RCT) [4] conducted in 2016 demonstrated that administering glucocorticoids to mothers at risk of late preterm labor substantially lowered the incidence of respiratory problems in newborns. On the basis of these findings—particularly the RCT evidence supporting ACS benefits in late preterm labor—clinical guidelines in many countries have been updated to include this patient population.

The timing of ACS administration is critical. Indeed, for early preterm birth, evidence suggests that the clinical benefits may diminish beyond 7 days following administration, leading to a consensus that the optimal benefit window is 2 to 7 days after the first dose [5–7]. A limitation of this landmark RCT [4], however, is that it did not analyze the connection between newborn outcomes and the time from ACS administration to delivery. Consequently, the optimal administration window for ACS in late preterm birth has not been established. This study was designed to evaluate how different administration-to-delivery intervals are associated with neonatal benefits and risks, thereby providing evidence to inform clinical timing decisions.

Methods

Study population

The study population included pregnant mothers admitted to the hospital’s obstetrics department between September 2021 and January 2025 (at 34^+ 0 to 36^+ 6 weeks of gestation) and their newborns. The inclusion criteria were as follows: (1) singleton pregnancy, (2) hospitalization in the obstetrics department within the aforementioned gestational age range, and (3) delivery at the hospital. The exclusion criteria included (1) multiple pregnancies, (2) the use of glucocorticoid therapy during pregnancy for maternal complications, (3) intrauterine fetal death, (4) maternal use of ACS prior to 34 weeks of gestation, and (5) maternal use of ACS outside the hospital with unknown details.

Data collection

Maternal and neonatal data were extracted from electronic medical records. Maternal variables comprised parity, age, mode of delivery, gestational age at delivery, and pregnancy complications, including hypertensive disorders during pregnancy (HDP), hyperglycemia, and premature rupture of membranes (PROM). Neonatal variables included sex, birthweight, Apgar score, admission to the neonatal intensive care unit (NICU), occurrence of neonatal respiratory distress syndrome (NRDS), transient tachypnea of the newborn (TTN), administration of pulmonary surfactant, type of respiratory support, and blood glucose levels.

Blood glucose monitoring for newborns was implemented selectively, not universally, across the cohort. Systematic screening was reserved for infants admitted to the NICU. Therefore, glucose data were available exclusively for this subset of newborns.

ACS treatment protocol

The ACS treatment protocol consisted of intramuscular injections of dexamethasone (6 mg every 12 h for a total of four doses) or betamethasone (12 mg every 24 h for a total of two doses). The ACS group comprised women who received at least one dose of either corticosteroid.

Study outcomes

The primary outcome was the incidence of NRDS. Neonatal hypoglycemia was assessed as a key secondary outcome. Other secondary outcomes included TTN, use of pulmonary surfactant, and the type of respiratory support required.

Definition of outcomes

NRDS was characterized by early-onset respiratory distress (e.g., tachypnea, grunting, cyanosis, and inspiratory retractions), with diagnosis confirmed through chest X-rays, which revealed a distinctive ground-glass pattern and air bronchogram signs. Neonates with a serum glucose concentration of 2.2 mmol/L or lower within 24 h after delivery were considered to have neonatal hypoglycemia. TTN was defined as the onset of tachypnea within 6 h of birth that may be accompanied by grunting, nasal flaring, mild cyanosis, or retractions, and is typically supported by chest X-ray showing mild bilateral pulmonary opacity. The pulmonary surfactant used in this study was injectable bovine lung surfactant. The types of respiratory support used included high-flow nasal cannula, continuous positive airway pressure, noninvasive positive pressure ventilation, and mechanical ventilation.

Statistical analyses

Patients were initially divided into ACS and non-ACS groups. The ACS group was further categorized based on the interval from the first ACS administration to delivery into three subgroups: less than 2 days, 2 to 7 days, and more than 7 days. Each ACS subgroup was then compared with the non-ACS group (reference) for all outcomes. For neonatal hypoglycemia, infants without recorded blood glucose measurements were considered not to have hypoglycemia in the primary analysis; a sensitivity analysis restricted to neonates admitted to the NICU was performed to test the robustness of this approach.

Continuous variables were reported as the means ± standard deviations or medians with interquartile ranges. For two-group comparisons, the independent samples t test or the Mann‒Whitney U test was used. One-way ANOVA or the Kruskal‒Wallis test was applied for comparisons involving more than two groups. The chi-square test or Fisher’s exact test was applied to compare the categorical variables, which were summarized as frequencies (n [%]). Variables that had a P value less than 0.2 in the univariable analyses were included in the multivariable logistic regression models to adjust for potential confounders for each outcome. A two-tailed P value < 0.05 was considered statistically significant. SPSS version 26.0 was used to conduct the statistical analyses.

Results

Study grouping process

Between September 2021 and January 2025, a total of 1538 pregnant women at 34^+ 0 to 36^+ 6 weeks of gestation were admitted to the obstetrics department of the hospital. Among these, 1113 gave birth in the hospital. According to the exclusion criteria, we excluded 115 cases of twin pregnancies, 14 women who received glucocorticoids due to complications, 7 cases of intrauterine fetal death, 128 women who received ACS before 34 weeks, and 61 women who received ACS outside the hospital (with unavailable details). Ultimately, 788 participants were enrolled in the study, with 537 (68.1%) having received ACS and 251 (31.9%) not receiving it. Among the 537 women who received ACS, 299 (55.7%) gave birth within 2 days of receiving ACS, 157 (29.2%) delivered between 2 and 7 days later, and 81 (15.1%) gave birth more than 7 days after receiving ACS (Fig. 1).

Fig. 1.

Flowchart of participant enrollment and grouping in the study of antenatal corticosteroid administration in pregnant women. ACS, antenatal corticosteroid

Comparison between the ACS and non-ACS groups

There were notable differences in the proportions of primiparas and the gestational age at delivery between the ACS and reference groups. Other clinical features, including the Apgar score, newborn sex, birthweight, and rate of NICU admission, as well as maternal age, mode of delivery, incidence of hyperglycemia during pregnancy, HDP, and PROM, were similar (Table 1).

Table 1.

Comparison of Clinical Characteristics and Neonatal Outcomes

Clinical Characteristics	Non-ACS group (n = 251)	ACS group
		Total (n = 537)	P	< 2d (n = 299)	2–7d (n = 157)	> 7d (n = 81)	P
Mother Characteristics
Age (years)	32.0(28.0, 35.0)	31.0(28.0, 34.0)	0.183	31.0(28.0, 34.0)	32.0(29.0, 35.0)	30.0(27.0, 33.0)	0.026
Primipara, n (%)	33.1%	40.8%	0.038	41.5%	31.8%	55.6%	0.001
C-Section, n (%)	85.7%	80.8%	0.097	76.6%	95.5%	67.9%	< 0.001
Hyperglycemia in pregnancy, n (%)	32.7%	28.9%	0.278	30.4%	28.0%	24.7%	0.522
HDP, n (%)	22.3%	21.2%	0.731	20.7%	25.5%	14.8%	0.283
PROM, n (%)	20.7%	22.7%	0.528	35.8%	7.6%	3.7%	< 0.001
Gestational age at ACS (weeks)	NA	35.86(34.86, 36.43)		35.71(34.86, 36.43)	36.00(34.86, 36.57)	35.57(34.86, 36.36)	0.674
Gestational age at delivery (weeks)	36.57(35.86, 37.00)	36.29(35.29, 37.00)	0.038	35.86(35.00, 36.43)	36.57(35.57, 37.00)	38.00(37.29–39.36)	< 0.001
Neonatal Characteristics
Male newborns, n (%)	56.6%	49.7%	0.073	49.8%	54.1%	40.7%	0.070
Birthweight (g)	2730(2430, 3060)	2740(2430, 3080)	0.648	2670(2340, 2920)	2760(2460, 3095)	3090(2760, 3395)	< 0.001
Apgar score (1-minute)	8.00(8.00, 9.00)	8.00(7.00, 9.00)	0.357	8.00(7.00, 8.00)	8.00(7.00, 9.00)	9.00(8.00, 9.00)	< 0.001
Apgar score (5-minute)	9.00(8.00, 10.00)	9.00(8.00, 10.00)	0.230	9.00(8.00, 9.00)	9.00(8.00, 10.00)	10.00(9.00, 10.00)	< 0.001
NICU, n (%)	48.6%	49.9%	0.734	59.9%	48.4%	16.0%	< 0.001
Neonatal Outcomes
NRDS, n (%)	25.5%	25.9%	0.908	33.4%	19.7%	9.9%	< 0.001
Hypoglycemia, n (%)	4.8%	12.7%	0.001	15.4%	12.7%	2.5%	< 0.001
TTN, n (%)	9.2%	9.1%	0.986	10.0%	11.5%	1.2%	0.060
HFNC/CPAP/NIPPV, n (%)	25.1%	28.1%	0.375	33.4%	27.4%	9.9%	< 0.001
MV, n (%)	9.2%	6.7%	0.222	8.7%	5.1%	2.5%	0.117
Lung surfactant, n (%)	8.8%	7.8%	0.651	10.4%	5.7%	2.5%	0.078

ACS Antenatal corticosteroid, PROM Premature rupture of membranes, NICU Neonatal intensive care unit, HDP Hypertensive disorders of pregnancy, TTN Transitory tachypnea of the newborn, NRDS Neonatal respiratory distress syndrome, HFNC High-flow nasal cannula, CPAP Continuous positive airway pressure, NIPPV Noninvasive positive pressure ventilation, MV Mechanical ventilation, NA Not applicable 

The overall incidence of NRDS did not differ significantly between the ACS and non-ACS groups. Nor were significant differences observed in the rates of TTN, use of pulmonary surfactant, or requirement for respiratory support. In contrast, the rate of neonatal hypoglycemia was significantly higher in the ACS group (12.7%) than in the non-ACS group (4.8%; P = 0.001). After multivariable logistic regression was used to control for potential confounders (defined as those with a P value < 0.2 in the univariable analysis, including maternal age, primiparity, cesarean delivery, gestational age at birth, and newborn sex), the ACS group continued to exhibit a substantially greater risk of neonatal hypoglycemia (odds ratio (OR) = 2.594, 95% confidence interval (CI): 1.359–4.951, P = 0.004) (Fig. 2).

Fig. 2.

Multivariable logistic regression analysis examining the relationship between the antenatal corticosteroid administration-to-delivery interval and neonatal outcomes. ACS, antenatal corticosteroid; TTN, transitory tachypnea of the newborn; NRDS, neonatal respiratory distress syndrome; HFNC, high-flow nasal cannula oxygen therapy; CPAP, continuous positive airway pressure; NIPPV, noninvasive positive pressure ventilation; MV, mechanical ventilation; OR, odds ratio; Cl, confidence interval. ^a adjusted for cesarean section, primiparity, maternal age, PROM, neonatal sex, birthweight, and gestational age at delivery. ^b Adjusted for cesarean section, primiparity, maternal age, neonatal sex, and gestational age at delivery

Subgroup analyses

Significant differences were observed among the < 2 days, 2–7 days, > 7 days, and non-ACS groups regarding maternal age, proportion of primiparas, cesarean section rates, occurrence of PROM, Apgar score, birthweight, gestational age at delivery, and frequency of NICU admissions. Neonatal sex and the occurrence of hyperglycemia during pregnancy and HDP were comparable across all four groups (Table 1).

The rates of NRDS and neonatal hypoglycemia, along with the type of respiratory support required, varied significantly across the groups (P < 0.05). The variables that had a P value < 0.2 in the univariate analysis (including maternal age, primiparity, cesarean section, PROM, neonatal sex, birthweight, and gestational age at delivery) were controlled for in the multivariable logistic regression models. Compared with the non-ACS group, only the 2–7 days group had a significantly lower risk of NRDS (OR = 0.468, 95% CI: 0.261–0.838, P = 0.011), while no significant association was observed for the < 2 days or > 7 days subgroups. In contrast, for neonatal hypoglycemia, a significantly higher risk was observed in both the < 2 days (OR = 2.922, 95% CI: 1.459–5.853, P = 0.002) and 2–7 days subgroups (OR = 2.641, 95% CI: 1.228–5.681, P = 0.013), while no increased risk was observed in the > 7 days subgroup. Regarding other outcomes, only the 2–7 days subgroup showed a reduced need for mechanical ventilation (OR = 0.365, 95% CI: 0.152–0.873, P = 0.024), whereas no significant differences across subgroups were observed for TTN or pulmonary surfactant use (Fig. 2).

Sensitivity analysis for neonatal hypoglycemia

To address potential detection bias from non-universal glucose screening, a sensitivity analysis was performed in the subset of infants admitted to the NICU (n = 390), all of whom underwent systematic monitoring. In this cohort, the incidence of hypoglycemia remained significantly higher in the ACS group than in the non-ACS group (24.6% vs. 9.8%; OR = 3.505, 95% CI: 1.749–7.025, P < 0.001). This elevated risk persisted when the ACS group was stratified by administration-to-delivery interval, with significant associations in both the < 2 days (OR = 3.829, 95% CI: 1.818–8.065, P < 0.001) and 2–7 days subgroups (OR = 3.502, 95% CI: 1.547–7.926, P = 0.003), but not in the > 7 days subgroup (Supplementary Table 1, Supplementary Fig. 1). These consistent findings confirm the robustness of the primary association between ACS use and neonatal hypoglycemia.

Discussion

Currently, there is a consensus that ACS should be given to improve preterm outcomes in pregnant mothers at risk of early preterm labor, but there remains controversy regarding whether ACS should be recommended for those at risk of late preterm labor between 34^+ 0 and 36^+ 6 weeks gestation. Furthermore, it is yet unknown how the timing of ACS administration affects late preterm infants. Our single-center historical cohort research demonstrated that in pregnant mothers at risk of late preterm birth, delivery within 2–7 days after ACS administration was linked to a lower risk of NRDS. Additionally, if delivery took place < 2 days or within 2–7 days following ACS injection, the risk of neonatal hypoglycemia increased.

In 2016, Gyamfi-Bannerman et al. conducted an RCT [4], with the primary outcome being neonatal death within 72 h, stillbirth, or neonatal receipt of respiratory support within the first 72 h (e.g., supplemental oxygen ≥ 30% within 4 h, continuous positive airway pressure or high-flow nasal cannula oxygen therapy for ≥ 2 h, mechanical ventilation, or extracorporeal membrane oxygenation). According to this study, whereas ACS did not significantly lower the incidence of NRDS, it did considerably reduce the incidence of certain infant respiratory complications. A systematic review including six RCTs [8] revealed that the ACS group had significantly lower risks of NRDS, TTN, severe NRDS, pulmonary surfactant usage, and mechanical ventilation. However, in our current investigation, the frequency of NRDS in the ACS and non-ACS groups did not differ statistically.

For neonates born to women at risk of early preterm birth, outcomes are known to be correlated with the ACS-to-delivery interval. Battarbee et al. [5] analyzed the best time to administer ACS for early preterm birth, and the results showed that glucocorticoids can enhance surfactant gene transcription in human lung cells, with the stimulatory effect peaking within 48 h and returning to near-baseline levels by day 8. Preterm newborns who receive ACS 2–7 days prior to delivery have the lowest risk of developing NRDS compared with those exposed to ACS either sooner or later than this time frame [5]. However, ACS has been widely used in clinical practice for late preterm infants in recent years. Therefore, how the ACS-to-delivery interval affects the outcomes of late preterm neonates is yet unknown. Gulersen et al.’s cohort study [9] revealed that infants born less than 2 days after ACS administration had a significantly higher risk of NRDS (OR = 9.86, 95% CI: 1.15–84.24, P = 0.04) than infants born within 2–7 days. Our results demonstrated that the incidence of NRDS was significantly lower in neonates born within 2–7 days following ACS administration than in the non-ACS group (OR = 0.468, 95% CI: 0.261–0.838, P = 0.011), whereas no significant difference was observed between neonates born < 2 days or > 7 days following ACS administration. The findings from these two studies, though interpreted from distinct perspectives, consistently indicate that the 2–7 day interval following ACS administration is associated with favorable respiratory outcomes in late preterm infants. This evidence highlights a clinically important time window that warrants further investigation to determine its potential role in optimizing neonatal care.

Prenatal glucocorticoid exposure may lead to neonatal hypoglycemia via two primary mechanisms: first, prenatal corticosteroid administration may induce maternal hyperglycemia, which promotes fetal pancreatic β-cell proliferation and subsequent hyperinsulinemia—ultimately resulting in neonatal hypoglycemia [10]; second, high doses of corticosteroids can cross the placental barrier and significantly suppress the fetal adrenal glands [11, 12]. Consistent with these mechanisms, clinical studies have revealed a link between ACS and neonatal hypoglycemia. For example, our results (12.7% vs. 4.8%; OR = 2.594, 95% CI: 1.359–4.951, P = 0.004) were in line with those of Gyamfi-Bannerman et al.'s study [4], which demonstrated that the incidence of neonatal hypoglycemia was greater in the ACS group than in the non-ACS group (24.0% vs. 14.9%; RR = 1.61, 95% CI: 1.38–1.88, P < 0.001). Notably, the incidence of neonatal hypoglycemia in our cohort was lower than that reported in the RCT, likely due to selective glucose screening limited to NICU-admitted infants. This interpretation is strongly supported by our sensitivity analysis within the NICU subgroup. The analysis not only reaffirmed a significant association between ACS use and hypoglycemia (OR = 3.505, 95% CI: 1.749–7.025, P < 0.001) but also yielded an incidence (24.6% vs. 9.8%) comparable to the RCT findings.

Regarding the effect of the ACS administration-to-delivery interval on neonatal hypoglycemia, Zigron et al. [13] reported that the only independent risk factor for hypoglycemia in early preterm infants was delivery within 24–48 h following ACS treatment. Gulersen et al.'s cohort study in late preterm birth reported that neonates delivered < 2 days after ACS had the highest risk of hypoglycemia (OR = 3.44, 95% CI: 2.10–5.63, P < 0.01), with the risk decreasing as the ACS-to-delivery interval lengthened (OR = 0.32, 95% CI: 0.20–0.51, P < 0.01 for the >7d group) [9]. However, that study did not adjust for gestational age at delivery, so the results cannot rule out the influence of gestational age on neonatal hypoglycemia. In contrast, our study, which adjusted for gestational age and other confounders, found that the increased risk of hypoglycemia remained significant in both the < 2 days (OR = 2.922, 95% CI: 1.459–5.853, P = 0.002) and 2–7 days subgroups (OR = 2.641, 95% CI: 1.228–5.681, P = 0.013), but was no longer apparent in the > 7 days subgroup. These results were confirmed in the sensitivity analysis. Therefore, routine blood glucose checks are advised for late preterm infants born to mothers who have been given ACS—especially those delivered within 7 days following ACS treatment—regardless of whether the neonates are admitted to the NICU.

Several limitations of this study should be considered. First, the study is not prospective; it is retrospective in nature. Consequently, the assessment of results depends on the availability of complete documentation in electronic medical records to ensure accurate data collection. Second, blood glucose screening was not universal but restricted to newborns admitted to the NICU. Consequently, the reported overall incidence of hypoglycemia is likely an underestimate. However, as our sensitivity analysis confined to the NICU-admitted subgroup demonstrated, this surveillance bias affected both groups equally (with comparable NICU admission rates) and did not alter the significant association between ACS use and hypoglycemia. Finally, the relatively limited sample size may affect the precision of the estimates and warrants validation in larger cohorts.

Conclusions

In conclusion, this retrospective analysis identified the ACS-to-delivery interval as a factor significantly associated with neonatal outcomes in late preterm birth. Delivery within 2–7 days was associated with reduced NRDS risk, but also with an increased risk of neonatal hypoglycemia when delivery occurred within 7 days of ACS. These associations underscore the need to monitor for and manage neonatal hypoglycemia after ACS administration and indicate that targeting this interval to improve respiratory outcomes should be evaluated in prospective studies.

Abbreviations

ACS

Antenatal corticosteroid

NRDS

Neonatal respiratory distress syndrome

TTN

Transient tachypnea of the newborn

OR

Odds ratio

CI

Confidence interval

RCT

Randomized controlled trial

NICU

Neonatal intensive care unit

HDP

Hypertensive disorders during pregnancy

PROM

Premature rupture of membranes

Authors’ contributions

All the authors contributed to the conception and design of the study. W.L. and T.W. conducted the data acquisition and data analysis. W.L. wrote the first draft of the manuscript. J.Z. revised the manuscript. All the authors read and approved the final manuscript.

Funding

This research was supported by two projects of the Jiangsu Pharmaceutical Association: (1) Hengrui Hospital Pharmaceutical Fund Research Project (Project Approval Number: H202346); (2) “Pharm” Research New Voice Pharmaceutical Research Project (Project Approval Number: 202564053).

Data availability

The data of this study is available from the corresponding authors on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of the Affiliated Hospital of Xuzhou Medical University. The approval date was May 17, 2024, with reference number XYFK2024-KL239-01. This study is a retrospective study, and the Ethics Committee approved us to obtain verbal informed consent from all participants by telephone.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.